Search

Reasoning is a fundamental capability for solving complex multi-step problems, particularly in visual contexts where sequential step-wise understanding is essential. Existing approaches lack a comprehensive framework for evaluating visual reasoning and do not emphasize step-wise problem-solving. To this end, we propose a comprehensive framework for advancing step-by-step visual reasoning in large language models (LMMs) through three key contributions. First, we introduce a visual reasoning benchmark specifically designed to evaluate multi-step reasoning tasks. The benchmark presents a diverse set of challenges with eight different categories ranging from complex visual perception to scientific reasoning with over 4k reasoning steps in total, enabling robust evaluation of LLMs' abilities to perform accurate and interpretable visual reasoning across multiple steps. Second, we propose a novel metric that assesses visual reasoning quality at the granularity of individual steps, emphasizing both correctness and logical coherence. The proposed metric offers deeper insights into reasoning performance compared to traditional end-task accuracy metrics. Third, we present a new multimodal visual reasoning model, named LlamaV-o1, trained using a multi-step curriculum learning approach, where tasks are progressively organized to facilitate incremental skill acquisition and problem-solving. The proposed LlamaV-o1 is designed for multi-step reasoning and learns step-by-step through a structured training paradigm. Extensive experiments show that our LlamaV-o1 outperforms existing open-source models and performs favorably against close-source proprietary models. Compared to the recent Llava-CoT, our LlamaV-o1 achieves an average score of 67.3 with an absolute gain of 3.8\% across six benchmarks while being 5 times faster during inference scaling. Our benchmark, model, and code are publicly available., Comment: 15 pages, 5 Figures

3D multi-object tracking plays a critical role in autonomous driving by enabling the real-time monitoring and prediction of multiple objects' movements. Traditional 3D tracking systems are typically constrained by predefined object categories, limiting their adaptability to novel, unseen objects in dynamic environments. To address this limitation, we introduce open-vocabulary 3D tracking, which extends the scope of 3D tracking to include objects beyond predefined categories. We formulate the problem of open-vocabulary 3D tracking and introduce dataset splits designed to represent various open-vocabulary scenarios. We propose a novel approach that integrates open-vocabulary capabilities into a 3D tracking framework, allowing for generalization to unseen object classes. Our method effectively reduces the performance gap between tracking known and novel objects through strategic adaptation. Experimental results demonstrate the robustness and adaptability of our method in diverse outdoor driving scenarios. To the best of our knowledge, this work is the first to address open-vocabulary 3D tracking, presenting a significant advancement for autonomous systems in real-world settings. Code, trained models, and dataset splits are available publicly., Comment: 7 pages, 4 figures, 3 tables

Mathematical solvers use parametrized Optimization Problems (OPs) as inputs to yield optimal decisions. In many real-world settings, some of these parameters are unknown or uncertain. Recent research focuses on predicting the value of these unknown parameters using available contextual features, aiming to decrease decision regret by adopting end-to-end learning approaches. However, these approaches disregard prediction uncertainty and therefore make the mathematical solver susceptible to provide erroneous decisions in case of low-confidence predictions. We propose a novel framework that models prediction uncertainty with Bayesian Neural Networks (BNNs) and propagates this uncertainty into the mathematical solver with a Stochastic Programming technique. The differentiable nature of BNNs and differentiable mathematical solvers allow for two different learning approaches: In the Decoupled learning approach, we update the BNN weights to increase the quality of the predictions' distribution of the OP parameters, while in the Combined learning approach, we update the weights aiming to directly minimize the expected OP's cost function in a stochastic end-to-end fashion. We do an extensive evaluation using synthetic data with various noise properties and a real dataset, showing that decisions regret are generally lower (better) with both proposed methods.

Recent works on open-vocabulary 3D instance segmentation show strong promise, but at the cost of slow inference speed and high computation requirements. This high computation cost is typically due to their heavy reliance on 3D clip features, which require computationally expensive 2D foundation models like Segment Anything (SAM) and CLIP for multi-view aggregation into 3D. As a consequence, this hampers their applicability in many real-world applications that require both fast and accurate predictions. To this end, we propose a fast yet accurate open-vocabulary 3D instance segmentation approach, named Open-YOLO 3D, that effectively leverages only 2D object detection from multi-view RGB images for open-vocabulary 3D instance segmentation. We address this task by generating class-agnostic 3D masks for objects in the scene and associating them with text prompts. We observe that the projection of class-agnostic 3D point cloud instances already holds instance information; thus, using SAM might only result in redundancy that unnecessarily increases the inference time. We empirically find that a better performance of matching text prompts to 3D masks can be achieved in a faster fashion with a 2D object detector. We validate our Open-YOLO 3D on two benchmarks, ScanNet200 and Replica, under two scenarios: (i) with ground truth masks, where labels are required for given object proposals, and (ii) with class-agnostic 3D proposals generated from a 3D proposal network. Our Open-YOLO 3D achieves state-of-the-art performance on both datasets while obtaining up to $\sim$16$\times$ speedup compared to the best existing method in literature. On ScanNet200 val. set, our Open-YOLO 3D achieves mean average precision (mAP) of 24.7\% while operating at 22 seconds per scene. Code and model are available at github.com/aminebdj/OpenYOLO3D., Comment: ICLR 2025 (Oral)

Learning latent costs of transitions on graphs from trajectories demonstrations under various contextual features is challenging but useful for path planning. Yet, existing methods either oversimplify cost assumptions or scale poorly with the number of observed trajectories. This paper introduces DataSP, a differentiable all-to-all shortest path algorithm to facilitate learning latent costs from trajectories. It allows to learn from a large number of trajectories in each learning step without additional computation. Complex latent cost functions from contextual features can be represented in the algorithm through a neural network approximation. We further propose a method to sample paths from DataSP in order to reconstruct/mimic observed paths' distributions. We prove that the inferred distribution follows the maximum entropy principle. We show that DataSP outperforms state-of-the-art differentiable combinatorial solver and classical machine learning approaches in predicting paths on graphs.

Recent advancements in 3D perception systems have significantly improved their ability to perform visual recognition tasks such as segmentation. However, these systems still heavily rely on explicit human instruction to identify target objects or categories, lacking the capability to actively reason and comprehend implicit user intentions. We introduce a novel segmentation task known as reasoning part segmentation for 3D objects, aiming to output a segmentation mask based on complex and implicit textual queries about specific parts of a 3D object. To facilitate evaluation and benchmarking, we present a large 3D dataset comprising over 60k instructions paired with corresponding ground-truth part segmentation annotations specifically curated for reasoning-based 3D part segmentation. We propose a model that is capable of segmenting parts of 3D objects based on implicit textual queries and generating natural language explanations corresponding to 3D object segmentation requests. Experiments show that our method achieves competitive performance to models that use explicit queries, with the additional abilities to identify part concepts, reason about them, and complement them with world knowledge. Our source code, dataset, and trained models are available at https://github.com/AmrinKareem/PARIS3D., Comment: 14 pages

Scene understanding plays an essential role in enabling autonomous driving and maintaining high standards of performance and safety. To address this task, cameras and laser scanners (LiDARs) have been the most commonly used sensors, with radars being less popular. Despite that, radars remain low-cost, information-dense, and fast-sensing techniques that are resistant to adverse weather conditions. While multiple works have been previously presented for radar-based scene semantic segmentation, the nature of the radar data still poses a challenge due to the inherent noise and sparsity, as well as the disproportionate foreground and background. In this work, we propose a novel approach to the semantic segmentation of radar scenes using a multi-input fusion of radar data through a novel architecture and loss functions that are tailored to tackle the drawbacks of radar perception. Our novel architecture includes an efficient attention block that adaptively captures important feature information. Our method, TransRadar, outperforms state-of-the-art methods on the CARRADA and RADIal datasets while having smaller model sizes. https://github.com/YahiDar/TransRadar

Existing 3D instance segmentation methods typically assume that all semantic classes to be segmented would be available during training and only seen categories are segmented at inference. We argue that such a closed-world assumption is restrictive and explore for the first time 3D indoor instance segmentation in an open-world setting, where the model is allowed to distinguish a set of known classes as well as identify an unknown object as unknown and then later incrementally learning the semantic category of the unknown when the corresponding category labels are available. To this end, we introduce an open-world 3D indoor instance segmentation method, where an auto-labeling scheme is employed to produce pseudo-labels during training and induce separation to separate known and unknown category labels. We further improve the pseudo-labels quality at inference by adjusting the unknown class probability based on the objectness score distribution. We also introduce carefully curated open-world splits leveraging realistic scenarios based on inherent object distribution, region-based indoor scene exploration and randomness aspect of open-world classes. Extensive experiments reveal the efficacy of the proposed contributions leading to promising open-world 3D instance segmentation performance., Comment: Accepted at NeurIPS 2023

Knots are deeply entangled with every branch of science. One of the biggest open challenges in knot theory is to formalise a knot invariant that can unambiguously and efficiently distinguish any two knotted curves. Additionally, the conjecture that the geometrical embedding of a curve encodes information on its underlying topology is, albeit physically intuitive, far from proven. Here we attempt to tackle both these outstanding challenges by proposing a neural network (NN) approach that takes as input a geometric representation of a knotted curve and tries to make predictions of the curve's topology. Intriguingly, we discover that NNs trained with a so-called geometrical "local writhe" representation of a knot can distinguish curves that share one or many topological invariants and knot polynomials, such as mutant and composite knots, and can thus classify knotted curves more precisely than some knot polynomials. Additionally, we also show that our approach can be scaled up to classify all prime knots up to 10-crossings with more than 95\% accuracy. Finally, we show that our NNs can also be trained to solve knot localisation problems on open and closed curves. Our main discovery is that the pattern of "local writhe" is a potentially unique geometric signature of the underlying topology of a curve. We hope that our results will suggest new methods for quantifying generic entanglements in soft matter and even inform new topological invariants., Comment: Accepted in Soft Matter

The performance of perception systems developed for autonomous driving vehicles has seen significant improvements over the last few years. This improvement was associated with the increasing use of LiDAR sensors and point cloud data to facilitate the task of object detection and recognition in autonomous driving. However, LiDAR and camera systems show deteriorating performances when used in unfavorable conditions like dusty and rainy weather. Radars on the other hand operate on relatively longer wavelengths which allows for much more robust measurements in these conditions. Despite that, radar-centric data sets do not get a lot of attention in the development of deep learning techniques for radar perception. In this work, we consider the radar object detection problem, in which the radar frequency data is the only input into the detection framework. We further investigate the challenges of using radar-only data in deep learning models. We propose a transformers-based model, named RadarFormer, that utilizes state-of-the-art developments in vision deep learning. Our model also introduces a channel-chirp-time merging module that reduces the size and complexity of our models by more than 10 times without compromising accuracy. Comprehensive experiments on the CRUW radar dataset demonstrate the advantages of the proposed method. Our RadarFormer performs favorably against the state-of-the-art methods while being 2x faster during inference and requiring only one-tenth of their model parameters. The code associated with this paper is available at https://github.com/YahiDar/RadarFormer., Comment: 18 pages (with reference), 8 figures, submitted and accepted to SCIA2023

Object detection in 3D point clouds is a crucial task in a range of computer vision applications including robotics, autonomous cars, and augmented reality. This work addresses the object detection task in 3D point clouds using a highly efficient, surface-biased, feature extraction method (wang2022rbgnet), that also captures contextual cues on multiple levels. We propose a 3D object detector that extracts accurate feature representations of object candidates and leverages self-attention on point patches, object candidates, and on the global scene in 3D scene. Self-attention is proven to be effective in encoding correlation information in 3D point clouds by (xie2020mlcvnet). While other 3D detectors focus on enhancing point cloud feature extraction by selectively obtaining more meaningful local features (wang2022rbgnet) where contextual information is overlooked. To this end, the proposed architecture uses ray-based surface-biased feature extraction and multi-level context encoding to outperform the state-of-the-art 3D object detector. In this work, 3D detection experiments are performed on scenes from the ScanNet dataset whereby the self-attention modules are introduced one after the other to isolate the effect of self-attention at each level.

Existing deep learning-based 3D object detectors typically rely on the appearance of individual objects and do not explicitly pay attention to the rich contextual information of the scene. In this work, we propose Contextualized Multi-Stage Refinement for 3D Object Detection (CMR3D) framework, which takes a 3D scene as input and strives to explicitly integrate useful contextual information of the scene at multiple levels to predict a set of object bounding-boxes along with their corresponding semantic labels. To this end, we propose to utilize a context enhancement network that captures the contextual information at different levels of granularity followed by a multi-stage refinement module to progressively refine the box positions and class predictions. Extensive experiments on the large-scale ScanNetV2 benchmark reveal the benefits of our proposed method, leading to an absolute improvement of 2.0% over the baseline. In addition to 3D object detection, we investigate the effectiveness of our CMR3D framework for the problem of 3D object counting. Our source code will be publicly released., Comment: 5 figures, 10 pages including references

The success of the transformer architecture in natural language processing has recently triggered attention in the computer vision field. The transformer has been used as a replacement for the widely used convolution operators, due to its ability to learn long-range dependencies. This replacement was proven to be successful in numerous tasks, in which several state-of-the-art methods rely on transformers for better learning. In computer vision, the 3D field has also witnessed an increase in employing the transformer for 3D convolution neural networks and multi-layer perceptron networks. Although a number of surveys have focused on transformers in vision in general, 3D vision requires special attention due to the difference in data representation and processing when compared to 2D vision. In this work, we present a systematic and thorough review of more than 100 transformers methods for different 3D vision tasks, including classification, segmentation, detection, completion, pose estimation, and others. We discuss transformer design in 3D vision, which allows it to process data with various 3D representations. For each application, we highlight key properties and contributions of proposed transformer-based methods. To assess the competitiveness of these methods, we compare their performance to common non-transformer methods on 12 3D benchmarks. We conclude the survey by discussing different open directions and challenges for transformers in 3D vision. In addition to the presented papers, we aim to frequently update the latest relevant papers along with their corresponding implementations at: https://github.com/lahoud/3d-vision-transformers.

In recent years, significant progress has been achieved for 3D object detection on point clouds thanks to the advances in 3D data collection and deep learning techniques. Nevertheless, 3D scenes exhibit a lot of variations and are prone to sensor inaccuracies as well as information loss during pre-processing. Thus, it is crucial to design techniques that are robust against these variations. This requires a detailed analysis and understanding of the effect of such variations. This work aims to analyze and benchmark popular point-based 3D object detectors against several data corruptions. To the best of our knowledge, we are the first to investigate the robustness of point-based 3D object detectors. To this end, we design and evaluate corruptions that involve data addition, reduction, and alteration. We further study the robustness of different modules against local and global variations. Our experimental results reveal several intriguing findings. For instance, we show that methods that integrate Transformers at a patch or object level lead to increased robustness, compared to using Transformers at the point level.

Background and purpose: The delineation of the gross tumor volume (GTV) is a critical step for radiation therapy treatment planning. The delineation procedure is typically performed manually which exposes two major issues: cost and reproducibility. Delineation is a time-consuming process that is subject to inter- and intra-observer variability. While methods have been proposed to predict GTV contours, typical approaches ignore variability and therefore fail to utilize the valuable confidence information offered by multiple contours. Materials and methods: In this work we propose an automatic GTV contouring method for soft-tissue sarcomas from X-ray computed tomography (CT) images, using deep learning by integrating inter- and intra-observer variability in the learned model. Sixty-eight patients with soft tissue and bone sarcomas were considered in this evaluation, all underwent pre-operative CT imaging used to perform GTV delineation. Four radiation oncologists and radiologists performed three contouring trials each for all patients. We quantify variability by defining confidence levels based on the frequency of inclusion of a given voxel into the GTV and use a deep convolutional neural network to learn GTV confidence maps. Results: Results were compared to confidence maps from the four readers as well as ground-truth consensus contours established jointly by all readers. The resulting continuous Dice score between predicted and true confidence maps was 87% and the Hausdorff distance was 14 mm. Conclusion: Results demonstrate the ability of the proposed method to predict accurate contours while utilizing variability and as such it can be used to improve clinical workflow.

-The focus of this paper is targeted towards multi-cell 5G networks which are composed of HPNs (High Power Node) such as evolved NodeBs (HPNs) that control signaling and system broadcasting information and of simplified LPNs (Low Power Node) co-existing in the same operating area and sharing the scare radio resources. Consequently, greater emphasis is given to Inter-Cell Interference Coordination (ICIC) based on multi-resource management techniques that take in particular user association into account. Beside user association to HPNs, this paper takes also power control and scheduling into consideration. This addressed problem is remained largely unsolved, mainly due to its non-convex nature that makes the global optimal solution difficult to obtain. We address the user association challenge according to the two broadly adopted approaches in wireless networks: the network-centric approach where user association is allocated efficiently in a centralized fashion; and the user-centric approach where distributed allocation prevails for reduced complexity. The scheduling and HPNs power allocation are solved in a centralized fashion, in order to reach an optimal solution of the joint optimization problem.

Although well-known large-scale datasets, such as ImageNet, have driven image understanding forward, most of these datasets require extensive manual annotation and are thus not easily scalable. This limits the advancement of image understanding techniques. The impact of these large-scale datasets can be observed in almost every vision task and technique in the form of pre-training for initialization. In this work, we propose an easily scalable and self-supervised technique that can be used to pre-train any semantic RGB segmentation method. In particular, our pre-training approach makes use of automatically generated labels that can be obtained using depth sensors. These labels, denoted by HN-labels, represent different height and normal patches, which allow mining of local semantic information that is useful in the task of semantic RGB segmentation. We show how our proposed self-supervised pre-training with HN-labels can be used to replace ImageNet pre-training, while using 25x less images and without requiring any manual labeling. We pre-train a semantic segmentation network with our HN-labels, which resembles our final task more than pre-training on a less related task, e.g. classification with ImageNet. We evaluate on two datasets (NYUv2 and CamVid), and we show how the similarity in tasks is advantageous not only in speeding up the pre-training process, but also in achieving better final semantic segmentation accuracy than ImageNet pre-training

We propose a novel method for instance label segmentation of dense 3D voxel grids. We target volumetric scene representations, which have been acquired with depth sensors or multi-view stereo methods and which have been processed with semantic 3D reconstruction or scene completion methods. The main task is to learn shape information about individual object instances in order to accurately separate them, including connected and incompletely scanned objects. We solve the 3D instance-labeling problem with a multi-task learning strategy. The first goal is to learn an abstract feature embedding, which groups voxels with the same instance label close to each other while separating clusters with different instance labels from each other. The second goal is to learn instance information by densely estimating directional information of the instance's center of mass for each voxel. This is particularly useful to find instance boundaries in the clustering post-processing step, as well as, for scoring the segmentation quality for the first goal. Both synthetic and real-world experiments demonstrate the viability and merits of our approach. In fact, it achieves state-of-the-art performance on the ScanNet 3D instance segmentation benchmark.

We propose a real-time image fusion method using pre-trained neural networks. Our method generates a single image containing features from multiple sources. We first decompose images into a base layer representing large scale intensity variations, and a detail layer containing small scale changes. We use visual saliency to fuse the base layers, and deep feature maps extracted from a pre-trained neural network to fuse the detail layers. We conduct ablation studies to analyze our method's parameters such as decomposition filters, weight construction methods, and network depth and architecture. Then, we validate its effectiveness and speed on thermal, medical, and multi-focus fusion. We also apply it to multiple image inputs such as multi-exposure sequences. The experimental results demonstrate that our technique achieves state-of-the-art performance in visual quality, objective assessment, and runtime efficiency., Comment: 13 pages, 10 figures

We present a method to train self-binarizing neural networks, that is, networks that evolve their weights and activations during training to become binary. To obtain similar binary networks, existing methods rely on the sign activation function. This function, however, has no gradients for non-zero values, which makes standard backpropagation impossible. To circumvent the difficulty of training a network relying on the sign activation function, these methods alternate between floating-point and binary representations of the network during training, which is sub-optimal and inefficient. We approach the binarization task by training on a unique representation involving a smooth activation function, which is iteratively sharpened during training until it becomes a binary representation equivalent to the sign activation function. Additionally, we introduce a new technique to perform binary batch normalization that simplifies the conventional batch normalization by transforming it into a simple comparison operation. This is unlike existing methods, which are forced to the retain the conventional floating-point-based batch normalization. Our binary networks, apart from displaying advantages of lower memory and computation as compared to conventional floating-point and binary networks, also show higher classification accuracy than existing state-of-the-art methods on multiple benchmark datasets., Comment: 9 pages, 5 figures