Search

Emergent-scene safety is the key milestone for fully autonomous driving, and reliable on-time prediction is essential to maintain safety in emergency scenarios. However, these emergency scenarios are long-tailed and hard to collect, which restricts the system from getting reliable predictions. In this paper, we build a new dataset, which aims at the long-term prediction with the inconspicuous state variation in history for the emergency event, named the Extro-Spective Prediction (ESP) problem. Based on the proposed dataset, a flexible feature encoder for ESP is introduced to various prediction methods as a seamless plug-in, and its consistent performance improvement underscores its efficacy. Furthermore, a new metric named clamped temporal error (CTE) is proposed to give a more comprehensive evaluation of prediction performance, especially in time-sensitive emergency events of subseconds. Interestingly, as our ESP features can be described in human-readable language naturally, the application of integrating into ChatGPT also shows huge potential. The ESP-dataset and all benchmarks are released at https://dingrui-wang.github.io/ESP-Dataset/., Comment: Accepted by ICRA 2024 as Oral Presentation

Bird's eye view (BEV) representation has emerged as a dominant solution for describing 3D space in autonomous driving scenarios. However, objects in the BEV representation typically exhibit small sizes, and the associated point cloud context is inherently sparse, which leads to great challenges for reliable 3D perception. In this paper, we propose IS-Fusion, an innovative multimodal fusion framework that jointly captures the Instance- and Scene-level contextual information. IS-Fusion essentially differs from existing approaches that only focus on the BEV scene-level fusion by explicitly incorporating instance-level multimodal information, thus facilitating the instance-centric tasks like 3D object detection. It comprises a Hierarchical Scene Fusion (HSF) module and an Instance-Guided Fusion (IGF) module. HSF applies Point-to-Grid and Grid-to-Region transformers to capture the multimodal scene context at different granularities. IGF mines instance candidates, explores their relationships, and aggregates the local multimodal context for each instance. These instances then serve as guidance to enhance the scene feature and yield an instance-aware BEV representation. On the challenging nuScenes benchmark, IS-Fusion outperforms all the published multimodal works to date. Code is available at: https://github.com/yinjunbo/IS-Fusion., Comment: Accepted to CVPR 2024; Code: https://github.com/yinjunbo/IS-Fusion

Vehicle-to-Everything (V2X) collaborative perception has recently gained significant attention due to its capability to enhance scene understanding by integrating information from various agents, e.g., vehicles, and infrastructure. However, current works often treat the information from each agent equally, ignoring the inherent domain gap caused by the utilization of different LiDAR sensors of each agent, thus leading to suboptimal performance. In this paper, we propose DI-V2X, that aims to learn Domain-Invariant representations through a new distillation framework to mitigate the domain discrepancy in the context of V2X 3D object detection. DI-V2X comprises three essential components: a domain-mixing instance augmentation (DMA) module, a progressive domain-invariant distillation (PDD) module, and a domain-adaptive fusion (DAF) module. Specifically, DMA builds a domain-mixing 3D instance bank for the teacher and student models during training, resulting in aligned data representation. Next, PDD encourages the student models from different domains to gradually learn a domain-invariant feature representation towards the teacher, where the overlapping regions between agents are employed as guidance to facilitate the distillation process. Furthermore, DAF closes the domain gap between the students by incorporating calibration-aware domain-adaptive attention. Extensive experiments on the challenging DAIR-V2X and V2XSet benchmark datasets demonstrate DI-V2X achieves remarkable performance, outperforming all the previous V2X models. Code is available at https://github.com/Serenos/DI-V2X, Comment: aaai2024

Current successful methods of 3D scene perception rely on the large-scale annotated point cloud, which is tedious and expensive to acquire. In this paper, we propose Model2Scene, a novel paradigm that learns free 3D scene representation from Computer-Aided Design (CAD) models and languages. The main challenges are the domain gaps between the CAD models and the real scene's objects, including model-to-scene (from a single model to the scene) and synthetic-to-real (from synthetic model to real scene's object). To handle the above challenges, Model2Scene first simulates a crowded scene by mixing data-augmented CAD models. Next, we propose a novel feature regularization operation, termed Deep Convex-hull Regularization (DCR), to project point features into a unified convex hull space, reducing the domain gap. Ultimately, we impose contrastive loss on language embedding and the point features of CAD models to pre-train the 3D network. Extensive experiments verify the learned 3D scene representation is beneficial for various downstream tasks, including label-free 3D object salient detection, label-efficient 3D scene perception and zero-shot 3D semantic segmentation. Notably, Model2Scene yields impressive label-free 3D object salient detection with an average mAP of 46.08\% and 55.49\% on the ScanNet and S3DIS datasets, respectively. The code will be publicly available., Comment: arXiv admin note: substantial text overlap with arXiv:2203.10546

This paper addresses the problem of 3D referring expression comprehension (REC) in autonomous driving scenario, which aims to ground a natural language to the targeted region in LiDAR point clouds. Previous approaches for REC usually focus on the 2D or 3D-indoor domain, which is not suitable for accurately predicting the location of the queried 3D region in an autonomous driving scene. In addition, the upper-bound limitation and the heavy computation cost motivate us to explore a better solution. In this work, we propose a new multi-modal visual grounding task, termed LiDAR Grounding. Then we devise a Multi-modal Single Shot Grounding (MSSG) approach with an effective token fusion strategy. It jointly learns the LiDAR-based object detector with the language features and predicts the targeted region directly from the detector without any post-processing. Moreover, the image feature can be flexibly integrated into our approach to provide rich texture and color information. The cross-modal learning enforces the detector to concentrate on important regions in the point cloud by considering the informative language expressions, thus leading to much better accuracy and efficiency. Extensive experiments on the Talk2Car dataset demonstrate the effectiveness of the proposed methods. Our work offers a deeper insight into the LiDAR-based grounding task and we expect it presents a promising direction for the autonomous driving community.

Image instance segmentation is a fundamental research topic in autonomous driving, which is crucial for scene understanding and road safety. Advanced learning-based approaches often rely on the costly 2D mask annotations for training. In this paper, we present a more artful framework, LiDAR-guided Weakly Supervised Instance Segmentation (LWSIS), which leverages the off-the-shelf 3D data, i.e., Point Cloud, together with the 3D boxes, as natural weak supervisions for training the 2D image instance segmentation models. Our LWSIS not only exploits the complementary information in multimodal data during training, but also significantly reduces the annotation cost of the dense 2D masks. In detail, LWSIS consists of two crucial modules, Point Label Assignment (PLA) and Graph-based Consistency Regularization (GCR). The former module aims to automatically assign the 3D point cloud as 2D point-wise labels, while the latter further refines the predictions by enforcing geometry and appearance consistency of the multimodal data. Moreover, we conduct a secondary instance segmentation annotation on the nuScenes, named nuInsSeg, to encourage further research on multimodal perception tasks. Extensive experiments on the nuInsSeg, as well as the large-scale Waymo, show that LWSIS can substantially improve existing weakly supervised segmentation models by only involving 3D data during training. Additionally, LWSIS can also be incorporated into 3D object detectors like PointPainting to boost the 3D detection performance for free. The code and dataset are available at https://github.com/Serenos/LWSIS., Comment: AAAI2023

LiDAR-based 3D object detection is an indispensable task in advanced autonomous driving systems. Though impressive detection results have been achieved by superior 3D detectors, they suffer from significant performance degeneration when facing unseen domains, such as different LiDAR configurations, different cities, and weather conditions. The mainstream approaches tend to solve these challenges by leveraging unsupervised domain adaptation (UDA) techniques. However, these UDA solutions just yield unsatisfactory 3D detection results when there is a severe domain shift, e.g., from Waymo (64-beam) to nuScenes (32-beam). To address this, we present a novel Semi-Supervised Domain Adaptation method for 3D object detection (SSDA3D), where only a few labeled target data is available, yet can significantly improve the adaptation performance. In particular, our SSDA3D includes an Inter-domain Adaptation stage and an Intra-domain Generalization stage. In the first stage, an Inter-domain Point-CutMix module is presented to efficiently align the point cloud distribution across domains. The Point-CutMix generates mixed samples of an intermediate domain, thus encouraging to learn domain-invariant knowledge. Then, in the second stage, we further enhance the model for better generalization on the unlabeled target set. This is achieved by exploring Intra-domain Point-MixUp in semi-supervised learning, which essentially regularizes the pseudo label distribution. Experiments from Waymo to nuScenes show that, with only 10% labeled target data, our SSDA3D can surpass the fully-supervised oracle model with 100% target label. Our code is available at https://github.com/yinjunbo/SSDA3D., Comment: Accepted by AAAI 2023

3D object detection received increasing attention in autonomous driving recently. Objects in 3D scenes are distributed with diverse orientations. Ordinary detectors do not explicitly model the variations of rotation and reflection transformations. Consequently, large networks and extensive data augmentation are required for robust detection. Recent equivariant networks explicitly model the transformation variations by applying shared networks on multiple transformed point clouds, showing great potential in object geometry modeling. However, it is difficult to apply such networks to 3D object detection in autonomous driving due to its large computation cost and slow reasoning speed. In this work, we present TED, an efficient Transformation-Equivariant 3D Detector to overcome the computation cost and speed issues. TED first applies a sparse convolution backbone to extract multi-channel transformation-equivariant voxel features; and then aligns and aggregates these equivariant features into lightweight and compact representations for high-performance 3D object detection. On the highly competitive KITTI 3D car detection leaderboard, TED ranked 1st among all submissions with competitive efficiency., Comment: Accepted by AAAI 2023

We investigate transductive zero-shot point cloud semantic segmentation, where the network is trained on seen objects and able to segment unseen objects. The 3D geometric elements are essential cues to imply a novel 3D object type. However, previous methods neglect the fine-grained relationship between the language and the 3D geometric elements. To this end, we propose a novel framework to learn the geometric primitives shared in seen and unseen categories' objects and employ a fine-grained alignment between language and the learned geometric primitives. Therefore, guided by language, the network recognizes the novel objects represented with geometric primitives. Specifically, we formulate a novel point visual representation, the similarity vector of the point's feature to the learnable prototypes, where the prototypes automatically encode geometric primitives via back-propagation. Besides, we propose a novel Unknown-aware InfoNCE Loss to fine-grained align the visual representation with language. Extensive experiments show that our method significantly outperforms other state-of-the-art methods in the harmonic mean-intersection-over-union (hIoU), with the improvement of 17.8\%, 30.4\%, 9.2\% and 7.9\% on S3DIS, ScanNet, SemanticKITTI and nuScenes datasets, respectively. Codes are available (https://github.com/runnanchen/Zero-Shot-Point-Cloud-Segmentation), Comment: ACM MM 2023

In recent years, vision-centric Bird's Eye View (BEV) perception has garnered significant interest from both industry and academia due to its inherent advantages, such as providing an intuitive representation of the world and being conducive to data fusion. The rapid advancements in deep learning have led to the proposal of numerous methods for addressing vision-centric BEV perception challenges. However, there has been no recent survey encompassing this novel and burgeoning research field. To catalyze future research, this paper presents a comprehensive survey of the latest developments in vision-centric BEV perception and its extensions. It compiles and organizes up-to-date knowledge, offering a systematic review and summary of prevalent algorithms. Additionally, the paper provides in-depth analyses and comparative results on various BEV perception tasks, facilitating the evaluation of future works and sparking new research directions. Furthermore, the paper discusses and shares valuable empirical implementation details to aid in the advancement of related algorithms., Comment: project page at https://github.com/4DVLab/Vision-Centric-BEV-Perception; 22 pages, 15 figures

Previous works for LiDAR-based 3D object detection mainly focus on the single-frame paradigm. In this paper, we propose to detect 3D objects by exploiting temporal information in multiple frames, i.e., the point cloud videos. We empirically categorize the temporal information into short-term and long-term patterns. To encode the short-term data, we present a Grid Message Passing Network (GMPNet), which considers each grid (i.e., the grouped points) as a node and constructs a k-NN graph with the neighbor grids. To update features for a grid, GMPNet iteratively collects information from its neighbors, thus mining the motion cues in grids from nearby frames. To further aggregate the long-term frames, we propose an Attentive Spatiotemporal Transformer GRU (AST-GRU), which contains a Spatial Transformer Attention (STA) module and a Temporal Transformer Attention (TTA) module. STA and TTA enhance the vanilla GRU to focus on small objects and better align the moving objects. Our overall framework supports both online and offline video object detection in point clouds. We implement our algorithm based on prevalent anchor-based and anchor-free detectors. The evaluation results on the challenging nuScenes benchmark show the superior performance of our method, achieving the 1st on the leaderboard without any bells and whistles, by the time the paper is submitted.

Accurately detecting and tracking pedestrians in 3D space is challenging due to large variations in rotations, poses and scales. The situation becomes even worse for dense crowds with severe occlusions. However, existing benchmarks either only provide 2D annotations, or have limited 3D annotations with low-density pedestrian distribution, making it difficult to build a reliable pedestrian perception system especially in crowded scenes. To better evaluate pedestrian perception algorithms in crowded scenarios, we introduce a large-scale multimodal dataset,STCrowd. Specifically, in STCrowd, there are a total of 219 K pedestrian instances and 20 persons per frame on average, with various levels of occlusion. We provide synchronized LiDAR point clouds and camera images as well as their corresponding 3D labels and joint IDs. STCrowd can be used for various tasks, including LiDAR-only, image-only, and sensor-fusion based pedestrian detection and tracking. We provide baselines for most of the tasks. In addition, considering the property of sparse global distribution and density-varying local distribution of pedestrians, we further propose a novel method, Density-aware Hierarchical heatmap Aggregation (DHA), to enhance pedestrian perception in crowded scenes. Extensive experiments show that our new method achieves state-of-the-art performance for pedestrian detection on various datasets., Comment: accepted at CVPR2022

Promising performance has been achieved for visual perception on the point cloud. However, the current methods typically rely on labour-extensive annotations on the scene scans. In this paper, we explore how synthetic models alleviate the real scene annotation burden, i.e., taking the labelled 3D synthetic models as reference for supervision, the neural network aims to recognize specific categories of objects on a real scene scan (without scene annotation for supervision). The problem studies how to transfer knowledge from synthetic 3D models to real 3D scenes and is named Referring Transfer Learning (RTL). The main challenge is solving the model-to-scene (from a single model to the scene) and synthetic-to-real (from synthetic model to real scene's object) gap between the synthetic model and the real scene. To this end, we propose a simple yet effective framework to perform two alignment operations. First, physical data alignment aims to make the synthetic models cover the diversity of the scene's objects with data processing techniques. Then a novel \textbf{convex-hull regularized feature alignment} introduces learnable prototypes to project the point features of both synthetic models and real scenes to a unified feature space, which alleviates the domain gap. These operations ease the model-to-scene and synthetic-to-real difficulty for a network to recognize the target objects on a real unseen scene. Experiments show that our method achieves the average mAP of 46.08\% and 55.49\% on the ScanNet and S3DIS datasets by learning the synthetic models from the ModelNet dataset. Code will be publicly available.

Recently, there have been many advances in autonomous driving society, attracting a lot of attention from academia and industry. However, existing works mainly focus on cars, extra development is still required for self-driving truck algorithms and models. In this paper, we introduce an intelligent self-driving truck system. Our presented system consists of three main components, 1) a realistic traffic simulation module for generating realistic traffic flow in testing scenarios, 2) a high-fidelity truck model which is designed and evaluated for mimicking real truck response in real-world deployment, 3) an intelligent planning module with learning-based decision making algorithm and multi-mode trajectory planner, taking into account the truck's constraints, road slope changes, and the surrounding traffic flow. We provide quantitative evaluations for each component individually to demonstrate the fidelity and performance of each part. We also deploy our proposed system on a real truck and conduct real world experiments which shows our system's capacity of mitigating sim-to-real gap. Our code is available at https://github.com/InceptioResearch/IITS

In this paper, we present a large-scale detailed 3D face dataset, FaceScape, and the corresponding benchmark to evaluate single-view facial 3D reconstruction. By training on FaceScape data, a novel algorithm is proposed to predict elaborate riggable 3D face models from a single image input. FaceScape dataset releases $16,940$ textured 3D faces, captured from $847$ subjects and each with $20$ specific expressions. The 3D models contain the pore-level facial geometry that is also processed to be topologically uniform. These fine 3D facial models can be represented as a 3D morphable model for coarse shapes and displacement maps for detailed geometry. Taking advantage of the large-scale and high-accuracy dataset, a novel algorithm is further proposed to learn the expression-specific dynamic details using a deep neural network. The learned relationship serves as the foundation of our 3D face prediction system from a single image input. Different from most previous methods, our predicted 3D models are riggable with highly detailed geometry under different expressions. We also use FaceScape data to generate the in-the-wild and in-the-lab benchmark to evaluate recent methods of single-view face reconstruction. The accuracy is reported and analyzed on the dimensions of camera pose and focal length, which provides a faithful and comprehensive evaluation and reveals new challenges. The unprecedented dataset, benchmark, and code have been released at https://github.com/zhuhao-nju/facescape., Comment: Accepted to T-PAMI 2023; Extension of FaceScape(CVPR 2020); Code & data are available at https://github.com/zhuhao-nju/facescape

State-of-the-art methods for driving-scene LiDAR-based perception (including point cloud semantic segmentation, panoptic segmentation and 3D detection, \etc) often project the point clouds to 2D space and then process them via 2D convolution. Although this cooperation shows the competitiveness in the point cloud, it inevitably alters and abandons the 3D topology and geometric relations. A natural remedy is to utilize the 3D voxelization and 3D convolution network. However, we found that in the outdoor point cloud, the improvement obtained in this way is quite limited. An important reason is the property of the outdoor point cloud, namely sparsity and varying density. Motivated by this investigation, we propose a new framework for the outdoor LiDAR segmentation, where cylindrical partition and asymmetrical 3D convolution networks are designed to explore the 3D geometric pattern while maintaining these inherent properties. The proposed model acts as a backbone and the learned features from this model can be used for downstream tasks such as point cloud semantic and panoptic segmentation or 3D detection. In this paper, we benchmark our model on these three tasks. For semantic segmentation, we evaluate the proposed model on several large-scale datasets, \ie, SemanticKITTI, nuScenes and A2D2. Our method achieves the state-of-the-art on the leaderboard of SemanticKITTI (both single-scan and multi-scan challenge), and significantly outperforms existing methods on nuScenes and A2D2 dataset. Furthermore, the proposed 3D framework also shows strong performance and good generalization on LiDAR panoptic segmentation and LiDAR 3D detection., Comment: Accepted by TPAMI 2021; Source code at https://github.com/xinge008/Cylinder3D. arXiv admin note: substantial text overlap with arXiv:2011.10033

This paper presents a novel framework to recover \emph{detailed} avatar from a single image. It is a challenging task due to factors such as variations in human shapes, body poses, texture, and viewpoints. Prior methods typically attempt to recover the human body shape using a parametric-based template that lacks the surface details. As such resulting body shape appears to be without clothing. In this paper, we propose a novel learning-based framework that combines the robustness of the parametric model with the flexibility of free-form 3D deformation. We use the deep neural networks to refine the 3D shape in a Hierarchical Mesh Deformation (HMD) framework, utilizing the constraints from body joints, silhouettes, and per-pixel shading information. Our method can restore detailed human body shapes with complete textures beyond skinned models. Experiments demonstrate that our method has outperformed previous state-of-the-art approaches, achieving better accuracy in terms of both 2D IoU number and 3D metric distance., Comment: Accepted by TPAMI. arXiv admin note: substantial text overlap with arXiv:1904.10506

In Autonomous Driving (AD) systems, perception is both security and safety critical. Despite various prior studies on its security issues, all of them only consider attacks on camera- or LiDAR-based AD perception alone. However, production AD systems today predominantly adopt a Multi-Sensor Fusion (MSF) based design, which in principle can be more robust against these attacks under the assumption that not all fusion sources are (or can be) attacked at the same time. In this paper, we present the first study of security issues of MSF-based perception in AD systems. We directly challenge the basic MSF design assumption above by exploring the possibility of attacking all fusion sources simultaneously. This allows us for the first time to understand how much security guarantee MSF can fundamentally provide as a general defense strategy for AD perception. We formulate the attack as an optimization problem to generate a physically-realizable, adversarial 3D-printed object that misleads an AD system to fail in detecting it and thus crash into it. We propose a novel attack pipeline that addresses two main design challenges: (1) non-differentiable target camera and LiDAR sensing systems, and (2) non-differentiable cell-level aggregated features popularly used in LiDAR-based AD perception. We evaluate our attack on MSF included in representative open-source industry-grade AD systems in real-world driving scenarios. Our results show that the attack achieves over 90% success rate across different object types and MSF. Our attack is also found stealthy, robust to victim positions, transferable across MSF algorithms, and physical-world realizable after being 3D-printed and captured by LiDAR and camera devices. To concretely assess the end-to-end safety impact, we further perform simulation evaluation and show that it can cause a 100% vehicle collision rate for an industry-grade AD system., Comment: Accepted by IEEE S&P 2021

Domain adaptive semantic segmentation refers to making predictions on a certain target domain with only annotations of a specific source domain. Current state-of-the-art works suggest that performing category alignment can alleviate domain shift reasonably. However, they are mainly based on image-to-image adversarial training and little consideration is given to semantic variations of an object among images, failing to capture a comprehensive picture of different categories. This motivates us to explore a holistic representative, the semantic distribution from each category in source domain, to mitigate the problem above. In this paper, we present semantic distribution-aware contrastive adaptation algorithm that enables pixel-wise representation alignment under the guidance of semantic distributions. Specifically, we first design a pixel-wise contrastive loss by considering the correspondences between semantic distributions and pixel-wise representations from both domains. Essentially, clusters of pixel representations from the same category should cluster together and those from different categories should spread out. Next, an upper bound on this formulation is derived by involving the learning of an infinite number of (dis)similar pairs, making it efficient. Finally, we verify that SDCA can further improve segmentation accuracy when integrated with the self-supervised learning. We evaluate SDCA on multiple benchmarks, achieving considerable improvements over existing algorithms.The code is publicly available at https://github.com/BIT-DA/SDCA, Comment: 15 pages

We present a novel method for single image depth estimation using surface normal constraints. Existing depth estimation methods either suffer from the lack of geometric constraints, or are limited to the difficulty of reliably capturing geometric context, which leads to a bottleneck of depth estimation quality. We therefore introduce a simple yet effective method, named Adaptive Surface Normal (ASN) constraint, to effectively correlate the depth estimation with geometric consistency. Our key idea is to adaptively determine the reliable local geometry from a set of randomly sampled candidates to derive surface normal constraint, for which we measure the consistency of the geometric contextual features. As a result, our method can faithfully reconstruct the 3D geometry and is robust to local shape variations, such as boundaries, sharp corners and noises. We conduct extensive evaluations and comparisons using public datasets. The experimental results demonstrate our method outperforms the state-of-the-art methods and has superior efficiency and robustness.

3D perception using sensors under vehicle industrial standard is the rigid demand in autonomous driving. MEMS LiDAR emerges with irresistible trend due to its lower cost, more robust, and meeting the mass-production standards. However, it suffers small field of view (FoV), slowing down the step of its population. In this paper, we propose LEAD, i.e., LiDAR Extender for Autonomous Driving, to extend the MEMS LiDAR by coupled image w.r.t both FoV and range. We propose a multi-stage propagation strategy based on depth distributions and uncertainty map, which shows effective propagation ability. Moreover, our depth outpainting/propagation network follows a teacher-student training fashion, which transfers depth estimation ability to depth completion network without any scale error passed. To validate the LiDAR extension quality, we utilize a high-precise laser scanner to generate a ground-truth dataset. Quantitative and qualitative evaluations show that our scheme outperforms SOTAs with a large margin. We believe the proposed LEAD along with the dataset would benefit the community w.r.t depth researches.

Holistically understanding an object and its 3D movable parts through visual perception models is essential for enabling an autonomous agent to interact with the world. For autonomous driving, the dynamics and states of vehicle parts such as doors, the trunk, and the bonnet can provide meaningful semantic information and interaction states, which are essential to ensuring the safety of the self-driving vehicle. Existing visual perception models mainly focus on coarse parsing such as object bounding box detection or pose estimation and rarely tackle these situations. In this paper, we address this important autonomous driving problem by solving three critical issues. First, to deal with data scarcity, we propose an effective training data generation process by fitting a 3D car model with dynamic parts to vehicles in real images before reconstructing human-vehicle interaction (VHI) scenarios. Our approach is fully automatic without any human interaction, which can generate a large number of vehicles in uncommon states (VUS) for training deep neural networks (DNNs). Second, to perform fine-grained vehicle perception, we present a multi-task network for VUS parsing and a multi-stream network for VHI parsing. Third, to quantitatively evaluate the effectiveness of our data augmentation approach, we build the first VUS dataset in real traffic scenarios (e.g., getting on/out or placing/removing luggage). Experimental results show that our approach advances other baseline methods in 2D detection and instance segmentation by a big margin (over 8%). In addition, our network yields large improvements in discovering and understanding these uncommon cases. Moreover, we have released the source code, the dataset, and the trained model on Github (https://github.com/zongdai/EditingForDNN).

In this paper, we propose a novel approach to convert given speech audio to a photo-realistic speaking video of a specific person, where the output video has synchronized, realistic, and expressive rich body dynamics. We achieve this by first generating 3D skeleton movements from the audio sequence using a recurrent neural network (RNN), and then synthesizing the output video via a conditional generative adversarial network (GAN). To make the skeleton movement realistic and expressive, we embed the knowledge of an articulated 3D human skeleton and a learned dictionary of personal speech iconic gestures into the generation process in both learning and testing pipelines. The former prevents the generation of unreasonable body distortion, while the later helps our model quickly learn meaningful body movement through a few recorded videos. To produce photo-realistic and high-resolution video with motion details, we propose to insert part attention mechanisms in the conditional GAN, where each detailed part, e.g. head and hand, is automatically zoomed in to have their own discriminators. To validate our approach, we collect a dataset with 20 high-quality videos from 1 male and 1 female model reading various documents under different topics. Compared with previous SoTA pipelines handling similar tasks, our approach achieves better results by a user study., Comment: Accepted by ACCV 2020

To get clear street-view and photo-realistic simulation in autonomous driving, we present an automatic video inpainting algorithm that can remove traffic agents from videos and synthesize missing regions with the guidance of depth/point cloud. By building a dense 3D map from stitched point clouds, frames within a video are geometrically correlated via this common 3D map. In order to fill a target inpainting area in a frame, it is straightforward to transform pixels from other frames into the current one with correct occlusion. Furthermore, we are able to fuse multiple videos through 3D point cloud registration, making it possible to inpaint a target video with multiple source videos. The motivation is to solve the long-time occlusion problem where an occluded area has never been visible in the entire video. To our knowledge, we are the first to fuse multiple videos for video inpainting. To verify the effectiveness of our approach, we build a large inpainting dataset in the real urban road environment with synchronized images and Lidar data including many challenge scenes, e.g., long time occlusion. The experimental results show that the proposed approach outperforms the state-of-the-art approaches for all the criteria, especially the RMSE (Root Mean Squared Error) has been reduced by about 13%.

We present a novel approach to detect, segment, and reconstruct complete textured 3D models of vehicles from a single image for autonomous driving. Our approach combines the strengths of deep learning and the elegance of traditional techniques from part-based deformable model representation to produce high-quality 3D models in the presence of severe occlusions. We present a new part-based deformable vehicle model that is used for instance segmentation and automatically generate a dataset that contains dense correspondences between 2D images and 3D models. We also present a novel end-to-end deep neural network to predict dense 2D/3D mapping and highlight its benefits. Based on the dense mapping, we are able to compute precise 6-DoF poses and 3D reconstruction results at almost interactive rates on a commodity GPU. We have integrated these algorithms with an autonomous driving system. In practice, our method outperforms the state-of-the-art methods for all major vehicle parsing tasks: 2D instance segmentation by 4.4 points (mAP), 6-DoF pose estimation by 9.11 points, and 3D detection by 1.37. Moreover, we have released all of the source code, dataset, and the trained model on Github.

Current methods for trajectory prediction operate in supervised manners, and therefore require vast quantities of corresponding ground truth data for training. In this paper, we present a novel, label-free algorithm, AutoTrajectory, for trajectory extraction and prediction to use raw videos directly. To better capture the moving objects in videos, we introduce dynamic points. We use them to model dynamic motions by using a forward-backward extractor to keep temporal consistency and using image reconstruction to keep spatial consistency in an unsupervised manner. Then we aggregate dynamic points to instance points, which stand for moving objects such as pedestrians in videos. Finally, we extract trajectories by matching instance points for prediction training. To the best of our knowledge, our method is the first to achieve unsupervised learning of trajectory extraction and prediction. We evaluate the performance on well-known trajectory datasets and show that our method is effective for real-world videos and can use raw videos to further improve the performance of existing models.

Omnidirectional 360{\deg} camera proliferates rapidly for autonomous robots since it significantly enhances the perception ability by widening the field of view(FoV). However, corresponding 360{\deg} depth sensors, which are also critical for the perception system, are still difficult or expensive to have. In this paper, we propose a low-cost 3D sensing system that combines an omnidirectional camera with a calibrated projective depth camera, where the depth from the limited FoV can be automatically extended to the rest of the recorded omnidirectional image. To accurately recover the missing depths, we design an omnidirectional depth extension convolutional neural network(ODE-CNN), in which a spherical feature transform layer(SFTL) is embedded at the end of feature encoding layers, and a deformable convolutional spatial propagation network(D-CSPN) is appended at the end of feature decoding layers. The former resamples the neighborhood of each pixel in the omnidirectional coordination to the projective coordination, which reduces the difficulty of feature learning, and the later automatically finds a proper context to well align the structures in the estimated depths via CNN w.r.t. the reference image, which significantly improves the visual quality. Finally, we demonstrate the effectiveness of proposed ODE-CNN over the popular 360D dataset and show that ODE-CNN significantly outperforms (relatively 33% reduction in-depth error) other state-of-the-art (SoTA) methods., Comment: Accepted by ICRA 2020, 7 pages, 5 figures

In this paper, we propose a novel approach to reconstruct 3D human body shapes based on a sparse set of RGBD frames using a single RGBD camera. We specifically focus on the realistic settings where human subjects move freely during the capture. The main challenge is how to robustly fuse these sparse frames into a canonical 3D model, under pose changes and surface occlusions. This is addressed by our new framework consisting of the following steps. First, based on a generative human template, for every two frames having sufficient overlap, an initial pairwise alignment is performed; It is followed by a global non-rigid registration procedure, in which partial results from RGBD frames are collected into a unified 3D shape, under the guidance of correspondences from the pairwise alignment; Finally, the texture map of the reconstructed human model is optimized to deliver a clear and spatially consistent texture. Empirical evaluations on synthetic and real datasets demonstrate both quantitatively and qualitatively the superior performance of our framework in reconstructing complete 3D human models with high fidelity. It is worth noting that our framework is flexible, with potential applications going beyond shape reconstruction. As an example, we showcase its use in reshaping and reposing to a new avatar., Comment: Accepted by TMM

Despite the remarkable progresses made in deep-learning based depth map super-resolution (DSR), how to tackle real-world degradation in low-resolution (LR) depth maps remains a major challenge. Existing DSR model is generally trained and tested on synthetic dataset, which is very different from what would get from a real depth sensor. In this paper, we argue that DSR models trained under this setting are restrictive and not effective in dealing with real-world DSR tasks. We make two contributions in tackling real-world degradation of different depth sensors. First, we propose to classify the generation of LR depth maps into two types: non-linear downsampling with noise and interval downsampling, for which DSR models are learned correspondingly. Second, we propose a new framework for real-world DSR, which consists of four modules : 1) An iterative residual learning module with deep supervision to learn effective high-frequency components of depth maps in a coarse-to-fine manner; 2) A channel attention strategy to enhance channels with abundant high-frequency components; 3) A multi-stage fusion module to effectively re-exploit the results in the coarse-to-fine process; and 4) A depth refinement module to improve the depth map by TGV regularization and input loss. Extensive experiments on benchmarking datasets demonstrate the superiority of our method over current state-of-the-art DSR methods., Comment: accepted by Conference on Computer Vision and Pattern Recognition 2020

This paper reviews the CVPR 2019 challenge on Autonomous Driving. Baidu's Robotics and Autonomous Driving Lab (RAL) providing 150 minutes labeled Trajectory and 3D Perception dataset including about 80k lidar point cloud and 1000km trajectories for urban traffic. The challenge has two tasks in (1) Trajectory Prediction and (2) 3D Lidar Object Detection. There are more than 200 teams submitted results on Leaderboard and more than 1000 participants attended the workshop.

Existing LiDAR-based 3D object detectors usually focus on the single-frame detection, while ignoring the spatiotemporal information in consecutive point cloud frames. In this paper, we propose an end-to-end online 3D video object detector that operates on point cloud sequences. The proposed model comprises a spatial feature encoding component and a spatiotemporal feature aggregation component. In the former component, a novel Pillar Message Passing Network (PMPNet) is proposed to encode each discrete point cloud frame. It adaptively collects information for a pillar node from its neighbors by iterative message passing, which effectively enlarges the receptive field of the pillar feature. In the latter component, we propose an Attentive Spatiotemporal Transformer GRU (AST-GRU) to aggregate the spatiotemporal information, which enhances the conventional ConvGRU with an attentive memory gating mechanism. AST-GRU contains a Spatial Transformer Attention (STA) module and a Temporal Transformer Attention (TTA) module, which can emphasize the foreground objects and align the dynamic objects, respectively. Experimental results demonstrate that the proposed 3D video object detector achieves state-of-the-art performance on the large-scale nuScenes benchmark., Comment: Accepted to CVPR 2020. Code: https://github.com/yinjunbo/3DVID

In this paper, we present a large-scale detailed 3D face dataset, FaceScape, and propose a novel algorithm that is able to predict elaborate riggable 3D face models from a single image input. FaceScape dataset provides 18,760 textured 3D faces, captured from 938 subjects and each with 20 specific expressions. The 3D models contain the pore-level facial geometry that is also processed to be topologically uniformed. These fine 3D facial models can be represented as a 3D morphable model for rough shapes and displacement maps for detailed geometry. Taking advantage of the large-scale and high-accuracy dataset, a novel algorithm is further proposed to learn the expression-specific dynamic details using a deep neural network. The learned relationship serves as the foundation of our 3D face prediction system from a single image input. Different than the previous methods, our predicted 3D models are riggable with highly detailed geometry under different expressions. The unprecedented dataset and code will be released to public for research purpose., Comment: Accepted to CVPR 2020

Current popular online multi-object tracking (MOT) solutions apply single object trackers (SOTs) to capture object motions, while often requiring an extra affinity network to associate objects, especially for the occluded ones. This brings extra computational overhead due to repetitive feature extraction for SOT and affinity computation. Meanwhile, the model size of the sophisticated affinity network is usually non-trivial. In this paper, we propose a novel MOT framework that unifies object motion and affinity model into a single network, named UMA, in order to learn a compact feature that is discriminative for both object motion and affinity measure. In particular, UMA integrates single object tracking and metric learning into a unified triplet network by means of multi-task learning. Such design brings advantages of improved computation efficiency, low memory requirement and simplified training procedure. In addition, we equip our model with a task-specific attention module, which is used to boost task-aware feature learning. The proposed UMA can be easily trained end-to-end, and is elegant - requiring only one training stage. Experimental results show that it achieves promising performance on several MOT Challenge benchmarks., Comment: Accepted to CVPR 2020. Code: https://github.com/yinjunbo/UMA-MOT

Conventional absolute camera pose via a Perspective-n-Point (PnP) solver often assumes that the correspondences between 2D image pixels and 3D points are given. When the correspondences between 2D and 3D points are not known a priori, the task becomes the much more challenging blind PnP problem. This paper proposes a deep CNN model which simultaneously solves for both the 6-DoF absolute camera pose and 2D--3D correspondences. Our model comprises three neural modules connected in sequence. First, a two-stream PointNet-inspired network is applied directly to both the 2D image keypoints and the 3D scene points in order to extract discriminative point-wise features harnessing both local and contextual information. Second, a global feature matching module is employed to estimate a matchability matrix among all 2D--3D pairs. Third, the obtained matchability matrix is fed into a classification module to disambiguate inlier matches. The entire network is trained end-to-end, followed by a robust model fitting (P3P-RANSAC) at test time only to recover the 6-DoF camera pose. Extensive tests on both real and simulated data have shown that our method substantially outperforms existing approaches, and is capable of processing thousands of points a second with the state-of-the-art accuracy., Comment: A blind-PnP solver

Semantic segmentation is a challenging task that needs to handle large scale variations, deformations and different viewpoints. In this paper, we develop a novel network named Gated Path Selection Network (GPSNet), which aims to learn adaptive receptive fields. In GPSNet, we first design a two-dimensional multi-scale network - SuperNet, which densely incorporates features from growing receptive fields. To dynamically select desirable semantic context, a gate prediction module is further introduced. In contrast to previous works that focus on optimizing sample positions on the regular grids, GPSNet can adaptively capture free form dense semantic contexts. The derived adaptive receptive fields are data-dependent, and are flexible that can model different object geometric transformations. On two representative semantic segmentation datasets, i.e., Cityscapes, and ADE20K, we show that the proposed approach consistently outperforms previous methods and achieves competitive performance without bells and whistles.

State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel "domain normalization" approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) a trainable non-local graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep learning models (e.g. MC-CNN) fine-tuned with test-domain data.

Motivated by the need for photo-realistic simulation in autonomous driving, in this paper we present a video inpainting algorithm \emph{AutoRemover}, designed specifically for generating street-view videos without any moving objects. In our setup we have two challenges: the first is the shadow, shadows are usually unlabeled but tightly coupled with the moving objects. The second is the large ego-motion in the videos. To deal with shadows, we build up an autonomous driving shadow dataset and design a deep neural network to detect shadows automatically. To deal with large ego-motion, we take advantage of the multi-source data, in particular the 3D data, in autonomous driving. More specifically, the geometric relationship between frames is incorporated into an inpainting deep neural network to produce high-quality structurally consistent video output. Experiments show that our method outperforms other state-of-the-art (SOTA) object removal algorithms, reducing the RMSE by over $19\%$.

Depth Completion deals with the problem of converting a sparse depth map to a dense one, given the corresponding color image. Convolutional spatial propagation network (CSPN) is one of the state-of-the-art (SoTA) methods of depth completion, which recovers structural details of the scene. In this paper, we propose CSPN++, which further improves its effectiveness and efficiency by learning adaptive convolutional kernel sizes and the number of iterations for the propagation, thus the context and computational resources needed at each pixel could be dynamically assigned upon requests. Specifically, we formulate the learning of the two hyper-parameters as an architecture selection problem where various configurations of kernel sizes and numbers of iterations are first defined, and then a set of soft weighting parameters are trained to either properly assemble or select from the pre-defined configurations at each pixel. In our experiments, we find weighted assembling can lead to significant accuracy improvements, which we referred to as "context-aware CSPN", while weighted selection, "resource-aware CSPN" can reduce the computational resource significantly with similar or better accuracy. Besides, the resource needed for CSPN++ can be adjusted w.r.t. the computational budget automatically. Finally, to avoid the side effects of noise or inaccurate sparse depths, we embed a gated network inside CSPN++, which further improves the performance. We demonstrate the effectiveness of CSPN++on the KITTI depth completion benchmark, where it significantly improves over CSPN and other SoTA methods., Comment: Camera Ready Version. Accepted by AAAI 2020

Deep reinforcement learning has great potential to acquire complex, adaptive behaviors for autonomous agents automatically. However, the underlying neural network polices have not been widely deployed in real-world applications, especially in these safety-critical tasks (e.g., autonomous driving). One of the reasons is that the learned policy cannot perform flexible and resilient behaviors as traditional methods to adapt to diverse environments. In this paper, we consider the problem that a mobile robot learns adaptive and resilient behaviors for navigating in unseen uncertain environments while avoiding collisions. We present a novel approach for uncertainty-aware navigation by introducing an uncertainty-aware predictor to model the environmental uncertainty, and we propose a novel uncertainty-aware navigation network to learn resilient behaviors in the prior unknown environments. To train the proposed uncertainty-aware network more stably and efficiently, we present the temperature decay training paradigm, which balances exploration and exploitation during the training process. Our experimental evaluation demonstrates that our approach can learn resilient behaviors in diverse environments and generate adaptive trajectories according to environmental uncertainties., Comment: accepted to ICRA 2020

Adaptive inference is a promising technique to improve the computational efficiency of deep models at test time. In contrast to static models which use the same computation graph for all instances, adaptive networks can dynamically adjust their structure conditioned on each input. While existing research on adaptive inference mainly focuses on designing more advanced architectures, this paper investigates how to train such networks more effectively. Specifically, we consider a typical adaptive deep network with multiple intermediate classifiers. We present three techniques to improve its training efficacy from two aspects: 1) a Gradient Equilibrium algorithm to resolve the conflict of learning of different classifiers; 2) an Inline Subnetwork Collaboration approach and a One-for-all Knowledge Distillation algorithm to enhance the collaboration among classifiers. On multiple datasets (CIFAR-10, CIFAR-100 and ImageNet), we show that the proposed approach consistently leads to further improved efficiency on top of state-of-the-art adaptive deep networks.

In 2D/3D object detection task, Intersection-over-Union (IoU) has been widely employed as an evaluation metric to evaluate the performance of different detectors in the testing stage. However, during the training stage, the common distance loss (\eg, $L_1$ or $L_2$) is often adopted as the loss function to minimize the discrepancy between the predicted and ground truth Bounding Box (Bbox). To eliminate the performance gap between training and testing, the IoU loss has been introduced for 2D object detection in \cite{yu2016unitbox} and \cite{rezatofighi2019generalized}. Unfortunately, all these approaches only work for axis-aligned 2D Bboxes, which cannot be applied for more general object detection task with rotated Bboxes. To resolve this issue, we investigate the IoU computation for two rotated Bboxes first and then implement a unified framework, IoU loss layer for both 2D and 3D object detection tasks. By integrating the implemented IoU loss into several state-of-the-art 3D object detectors, consistent improvements have been achieved for both bird-eye-view 2D detection and point cloud 3D detection on the public KITTI benchmark., Comment: Accepted by international conference on 3d vision 2019

Deep neural networks (DNNs) are found to be vulnerable against adversarial examples, which are carefully crafted inputs with a small magnitude of perturbation aiming to induce arbitrarily incorrect predictions. Recent studies show that adversarial examples can pose a threat to real-world security-critical applications: a "physical adversarial Stop Sign" can be synthesized such that the autonomous driving cars will misrecognize it as others (e.g., a speed limit sign). However, these image-space adversarial examples cannot easily alter 3D scans of widely equipped LiDAR or radar on autonomous vehicles. In this paper, we reveal the potential vulnerabilities of LiDAR-based autonomous driving detection systems, by proposing an optimization based approach LiDAR-Adv to generate adversarial objects that can evade the LiDAR-based detection system under various conditions. We first show the vulnerabilities using a blackbox evolution-based algorithm, and then explore how much a strong adversary can do, using our gradient-based approach LiDAR-Adv. We test the generated adversarial objects on the Baidu Apollo autonomous driving platform and show that such physical systems are indeed vulnerable to the proposed attacks. We also 3D-print our adversarial objects and perform physical experiments to illustrate that such vulnerability exists in the real world. Please find more visualizations and results on the anonymous website: https://sites.google.com/view/lidar-adv.

This paper presents a novel framework to recover detailed human body shapes from a single image. It is a challenging task due to factors such as variations in human shapes, body poses, and viewpoints. Prior methods typically attempt to recover the human body shape using a parametric based template that lacks the surface details. As such the resulting body shape appears to be without clothing. In this paper, we propose a novel learning-based framework that combines the robustness of parametric model with the flexibility of free-form 3D deformation. We use the deep neural networks to refine the 3D shape in a Hierarchical Mesh Deformation (HMD) framework, utilizing the constraints from body joints, silhouettes, and per-pixel shading information. We are able to restore detailed human body shapes beyond skinned models. Experiments demonstrate that our method has outperformed previous state-of-the-art approaches, achieving better accuracy in terms of both 2D IoU number and 3D metric distance. The code is available in https://github.com/zhuhao-nju/hmd.git, Comment: CVPR 2019 Oral

As an essential problem in computer vision, salient object detection (SOD) has attracted an increasing amount of research attention over the years. Recent advances in SOD are predominantly led by deep learning-based solutions (named deep SOD). To enable in-depth understanding of deep SOD, in this paper, we provide a comprehensive survey covering various aspects, ranging from algorithm taxonomy to unsolved issues. In particular, we first review deep SOD algorithms from different perspectives, including network architecture, level of supervision, learning paradigm, and object-/instance-level detection. Following that, we summarize and analyze existing SOD datasets and evaluation metrics. Then, we benchmark a large group of representative SOD models, and provide detailed analyses of the comparison results. Moreover, we study the performance of SOD algorithms under different attribute settings, which has not been thoroughly explored previously, by constructing a novel SOD dataset with rich attribute annotations covering various salient object types, challenging factors, and scene categories. We further analyze, for the first time in the field, the robustness of SOD models to random input perturbations and adversarial attacks. We also look into the generalization and difficulty of existing SOD datasets. Finally, we discuss several open issues of SOD and outline future research directions., Comment: Published on IEEE TPAMI. All the saliency prediction maps, our constructed dataset with annotations, and codes for evaluation are publicly available at \url{https://github.com/wenguanwang/SODsurvey}

In the stereo matching task, matching cost aggregation is crucial in both traditional methods and deep neural network models in order to accurately estimate disparities. We propose two novel neural net layers, aimed at capturing local and the whole-image cost dependencies respectively. The first is a semi-global aggregation layer which is a differentiable approximation of the semi-global matching, the second is the local guided aggregation layer which follows a traditional cost filtering strategy to refine thin structures. These two layers can be used to replace the widely used 3D convolutional layer which is computationally costly and memory-consuming as it has cubic computational/memory complexity. In the experiments, we show that nets with a two-layer guided aggregation block easily outperform the state-of-the-art GC-Net which has nineteen 3D convolutional layers. We also train a deep guided aggregation network (GA-Net) which gets better accuracies than state-of-the-art methods on both Scene Flow dataset and KITTI benchmarks., Comment: CVPR 2019 (Oral Presentation)

Simulation systems have become an essential component in the development and validation of autonomous driving technologies. The prevailing state-of-the-art approach for simulation is to use game engines or high-fidelity computer graphics (CG) models to create driving scenarios. However, creating CG models and vehicle movements (e.g., the assets for simulation) remains a manual task that can be costly and time-consuming. In addition, the fidelity of CG images still lacks the richness and authenticity of real-world images and using these images for training leads to degraded performance. In this paper we present a novel approach to address these issues: Augmented Autonomous Driving Simulation (AADS). Our formulation augments real-world pictures with a simulated traffic flow to create photo-realistic simulation images and renderings. More specifically, we use LiDAR and cameras to scan street scenes. From the acquired trajectory data, we generate highly plausible traffic flows for cars and pedestrians and compose them into the background. The composite images can be re-synthesized with different viewpoints and sensor models. The resulting images are photo-realistic, fully annotated, and ready for end-to-end training and testing of autonomous driving systems from perception to planning. We explain our system design and validate our algorithms with a number of autonomous driving tasks from detection to segmentation and predictions. Compared to traditional approaches, our method offers unmatched scalability and realism. Scalability is particularly important for AD simulation and we believe the complexity and diversity of the real world cannot be realistically captured in a virtual environment. Our augmented approach combines the flexibility in a virtual environment (e.g., vehicle movements) with the richness of the real world to allow effective simulation of anywhere on earth.