Search

Diabetic retinopathy (DR) grading from fundus images has attracted increasing interest in both academic and industrial communities. Most convolutional neural network (CNN) based algorithms treat DR grading as a classification task via image-level annotations. However, these algorithms have not fully explored the valuable information in the DR-related lesions. In this paper, we present a robust framework, which collaboratively utilizes patch-level and image-level annotations, for DR severity grading. By an end-to-end optimization, this framework can bi-directionally exchange the fine-grained lesion and image-level grade information. As a result, it exploits more discriminative features for DR grading. The proposed framework shows better performance than the recent state-of-the-art algorithms and three clinical ophthalmologists with over nine years of experience. By testing on datasets of different distributions (such as label and camera), we prove that our algorithm is robust when facing image quality and distribution variations that commonly exist in real-world practice. We inspect the proposed framework through extensive ablation studies to indicate the effectiveness and necessity of each motivation. The code and some valuable annotations are now publicly available.

Image-text matching has received growing interest since it bridges vision and language. The key challenge lies in how to learn correspondence between image and text. Existing works learn coarse correspondence based on object co-occurrence statistics, while failing to learn fine-grained phrase correspondence. In this paper, we present a novel Graph Structured Matching Network (GSMN) to learn fine-grained correspondence. The GSMN explicitly models object, relation and attribute as a structured phrase, which not only allows to learn correspondence of object, relation and attribute separately, but also benefits to learn fine-grained correspondence of structured phrase. This is achieved by node-level matching and structure-level matching. The node-level matching associates each node with its relevant nodes from another modality, where the node can be object, relation or attribute. The associated nodes then jointly infer fine-grained correspondence by fusing neighborhood associations at structure-level matching. Comprehensive experiments show that GSMN outperforms state-of-the-art methods on benchmarks, with relative Recall@1 improvements of nearly 7% and 2% on Flickr30K and MSCOCO, respectively. Code will be released at: https://github.com/CrossmodalGroup/GSMN., Comment: Accepted by CVPR2020

With the ultimate goal of replacing proprietary hardware appliances with Virtual Network Functions (VNFs) implemented in software, Network Function Virtualization (NFV) has been gaining popularity in the past few years. Software switches route traffic between VNFs and physical Network Interface Cards (NICs). It is of paramount importance to compare the performance of different switch designs and architectures. In this paper, we propose a methodology to compare fairly and comprehensively the performance of software switches. We first explore the design spaces of seven state-of-the-art software switches and then compare their performance under four representative test scenarios. Each scenario corresponds to a specific case of routing NFV traffic between NICs and/or VNFs. In our experiments, we evaluate the throughput and latency between VNFs in two of the most popular virtualization environments, namely virtual machines (VMs) and containers. Our experimental results show that no single software switch prevails in all scenarios. It is, therefore, crucial to choose the most suitable solution for the given use case. At the same time, the presented results and analysis provide a deeper insight into the design tradeoffs and identifies potential performance bottlenecks that could inspire new designs., Comment: 17 pages

Cross-modality person re-identification (cm-ReID) is a challenging but key technology for intelligent video analysis. Existing works mainly focus on learning common representation by embedding different modalities into a same feature space. However, only learning the common characteristics means great information loss, lowering the upper bound of feature distinctiveness. In this paper, we tackle the above limitation by proposing a novel cross-modality shared-specific feature transfer algorithm (termed cm-SSFT) to explore the potential of both the modality-shared information and the modality-specific characteristics to boost the re-identification performance. We model the affinities of different modality samples according to the shared features and then transfer both shared and specific features among and across modalities. We also propose a complementary feature learning strategy including modality adaption, project adversarial learning and reconstruction enhancement to learn discriminative and complementary shared and specific features of each modality, respectively. The entire cm-SSFT algorithm can be trained in an end-to-end manner. We conducted comprehensive experiments to validate the superiority of the overall algorithm and the effectiveness of each component. The proposed algorithm significantly outperforms state-of-the-arts by 22.5% and 19.3% mAP on the two mainstream benchmark datasets SYSU-MM01 and RegDB, respectively., Comment: To appear at CVPR2020

Due to the intrinsically inefficient service provisioning in traditional networks, Network Function Virtualization (NFV) keeps gaining attention from both industry and academia. By replacing the purpose-built, expensive, proprietary network equipment with software network functions consolidated on commodity hardware, NFV envisions a shift towards a more agile and open service provisioning paradigm with much lower capital expenditure (CapEx) and operational expenditure (OpEx). Nonetheless, just like any complex system, NFV platforms commonly consist of abounding software and hardware components and usually incorporate disparate design choices based on distinct motivations or use cases. This broad collection of convoluted alternatives makes it extremely arduous for network operators to make proper choices. Although numerous efforts have been devoted to investigating different aspects of NFV, none of them specifically focused on NFV platforms or attempted to explore the design space. In this paper, we present a comprehensive survey on NFV platform design. Our study solely targets existing NFV platform implementations. We begin with an architectural view of the standard reference NFV platform and present our taxonomy of existing NFV platforms based on the principal purpose of design. Then we thoroughly explore the design space and elaborate on the implementation choices each platform opts for. We believe that our study gives a detailed guideline for network operators or service providers to choose or implement the most appropriate NFV platforms based on their respective requirements. Note that this report serves as a complementary document for a published IEEE TNSM paper [1]. We will periodically update this document to include the newly proposed NFV platforms and design choices., Comment: This report will be updated every six months to include the most up-to-date NFV platforms and design choices, in IEEE Transactions on Network and Service Management

RGB-Infrared (IR) person re-identification is an important and challenging task due to large cross-modality variations between RGB and IR images. Most conventional approaches aim to bridge the cross-modality gap with feature alignment by feature representation learning. Different from existing methods, in this paper, we propose a novel and end-to-end Alignment Generative Adversarial Network (AlignGAN) for the RGB-IR RE-ID task. The proposed model enjoys several merits. First, it can exploit pixel alignment and feature alignment jointly. To the best of our knowledge, this is the first work to model the two alignment strategies jointly for the RGB-IR RE-ID problem. Second, the proposed model consists of a pixel generator, a feature generator, and a joint discriminator. By playing a min-max game among the three components, our model is able to not only alleviate the cross-modality and intra-modality variations but also learn identity-consistent features. Extensive experimental results on two standard benchmarks demonstrate that the proposed model performs favorably against state-of-the-art methods. Especially, on SYSU-MM01 dataset, our model can achieve an absolute gain of 15.4% and 12.9% in terms of Rank-1 and mAP., Comment: accepted by ICCV'19

Learning semantic correspondence between image and text is significant as it bridges the semantic gap between vision and language. The key challenge is to accurately find and correlate shared semantics in image and text. Most existing methods achieve this goal by representing the shared semantic as a weighted combination of all the fragments (image regions or text words), where fragments relevant to the shared semantic obtain more attention, otherwise less. However, despite relevant ones contribute more to the shared semantic, irrelevant ones will more or less disturb it, and thus will lead to semantic misalignment in the correlation phase. To address this issue, we present a novel Bidirectional Focal Attention Network (BFAN), which not only allows to attend to relevant fragments but also diverts all the attention into these relevant fragments to concentrate on them. The main difference with existing works is they mostly focus on learning attention weight while our BFAN focus on eliminating irrelevant fragments from the shared semantic. The focal attention is achieved by pre-assigning attention based on inter-modality relation, identifying relevant fragments based on intra-modality relation and reassigning attention. Furthermore, the focal attention is jointly applied in both image-to-text and text-to-image directions, which enables to avoid preference to long text or complex image. Experiments show our simple but effective framework significantly outperforms state-of-the-art, with relative Recall@1 gains of 2.2% on both Flicr30K and MSCOCO benchmarks., Comment: Accepted by ACMMM2019

Temporal action localization has recently attracted significant interest in the Computer Vision community. However, despite the great progress, it is hard to identify which aspects of the proposed methods contribute most to the increase in localization performance. To address this issue, we conduct ablative experiments on feature extraction methods, fixed-size feature representation methods and training strategies, and report how each influences the overall performance. Based on our findings, we propose a two-stage detector that outperforms the state of the art in THUMOS14, achieving a mAP@tIoU=0.5 equal to 44.2%., Comment: ICIP19 Camera Ready

The tracking-by-detection framework requires a set of positive and negative training samples to learn robust tracking models for precise localization of target objects. However, existing tracking models mostly treat different samples independently while ignores the relationship information among them. In this paper, we propose a novel structure-aware deep neural network to overcome such limitations. In particular, we construct a graph to represent the pairwise relationships among training samples, and additionally take the natural language as the supervised information to learn both feature representations and classifiers robustly. To refine the states of the target and re-track the target when it is back to view from heavy occlusion and out of view, we elaborately design a novel subnetwork to learn the target-driven visual attentions from the guidance of both visual and natural language cues. Extensive experiments on five tracking benchmark datasets validated the effectiveness of our proposed method.

We consider a distributed Software Defined Networking (SDN) architecture adopting a cluster of multiple controllers to improve network performance and reliability. Besides the Openflow control traffic exchanged between controllers and switches, we focus on the control traffic exchanged among the controllers in the cluster, needed to run coordination and consensus algorithms to keep the controllers synchronized. We estimate the effect of the inter-controller communications on the reaction time perceived by the switches depending on the data-ownership model adopted in the cluster. The model is accurately validated in an operational Software Defined WAN (SDWAN). We advocate a careful placement of the controllers, that should take into account both the above kinds of control traffic. We evaluate, for some real ISP network topologies, the delay tradeoffs for the controllers placement problem and we propose a novel evolutionary algorithm to find the corresponding Pareto frontier. Our work provides novel quantitative tools to optimize the planning and the design of the network supporting the control plane of SDN networks, especially when the network is very large and in-band control plane is adopted. We also show that for operational distributed controllers (e.g. OpenDaylight and ONOS), the location of the controller which acts as a leader in the consensus algorithm has a strong impact on the reactivity perceived by switches., Comment: 14 pages