Bin Zhao

Zoey Guo, Yiwen Tang, Ray Zhang et al.

Understanding 3D scenes from multi-view inputs has been proven to alleviate the view discrepancy issue in 3D visual grounding. However, existing methods normally neglect the view cues embedded in the text modality and fail to weigh the relative importance of different views. In this paper, we propose ViewRefer, a multi-view framework for 3D visual grounding exploring how to grasp the view knowledge from both text and 3D modalities. For the text branch, ViewRefer leverages the diverse linguistic knowledge of large-scale language models, e.g., GPT, to expand a single grounding text to multiple geometry-consistent descriptions. Meanwhile, in the 3D modality, a transformer fusion module with inter-view attention is introduced to boost the interaction of objects across views. On top of that, we further present a set of learnable multi-view prototypes, which memorize scene-agnostic knowledge for different views, and enhance the framework from two perspectives: a view-guided attention module for more robust text features, and a view-guided scoring strategy during the final prediction. With our designed paradigm, ViewRefer achieves superior performance on three benchmarks and surpasses the second-best by +2.8%, +1.5%, and +1.35% on Sr3D, Nr3D, and ScanRefer. Code is released at https://github.com/Ivan-Tang-3D/ViewRefer3D.

Guanzhou Lan, Bin Zhao, Xuelong Li

Nighttime surveillance suffers from degradation due to poor illumination and arduous human annotations. It is challengable and remains a security risk at night. Existing methods rely on multi-spectral images to perceive objects in the dark, which are troubled by low resolution and color absence. We argue that the ultimate solution for nighttime surveillance is night-to-day translation, or Night2Day, which aims to translate a surveillance scene from nighttime to the daytime while maintaining semantic consistency. To achieve this, this paper presents a Disentangled Contrastive (DiCo) learning method. Specifically, to address the poor and complex illumination in the nighttime scenes, we propose a learnable physical prior, i.e., the color invariant, which provides a stable perception of a highly dynamic night environment and can be incorporated into the learning pipeline of neural networks. Targeting the surveillance scenes, we develop a disentangled representation, which is an auxiliary pretext task that separates surveillance scenes into the foreground and background with contrastive learning. Such a strategy can extract the semantics without supervision and boost our model to achieve instance-aware translation. Finally, we incorporate all the modules above into generative adversarial networks and achieve high-fidelity translation. This paper also contributes a new surveillance dataset called NightSuR. It includes six scenes to support the study on nighttime surveillance. This dataset collects nighttime images with different properties of nighttime environments, such as flare and extreme darkness. Extensive experiments demonstrate that our method outperforms existing works significantly. The dataset and source code will be released on GitHub soon.

Xianqiang Gao, Pingrui Zhang, Delin Qu et al.

3D Object Affordance Grounding aims to predict the functional regions on a 3D object and has laid the foundation for a wide range of applications in robotics. Recent advances tackle this problem via learning a mapping between 3D regions and a single human-object interaction image. However, the geometric structure of the 3D object and the object in the human-object interaction image are not always consistent, leading to poor generalization. To address this issue, we propose to learn generalizable invariant affordance knowledge from multiple human-object interaction images within the same affordance category. Specifically, we introduce the \textbf{M}ulti-\textbf{I}mage Guided Invariant-\textbf{F}eature-Aware 3D \textbf{A}ffordance \textbf{G}rounding (\textbf{MIFAG}) framework. It grounds 3D object affordance regions by identifying common interaction patterns across multiple human-object interaction images. First, the Invariant Affordance Knowledge Extraction Module (\textbf{IAM}) utilizes an iterative updating strategy to gradually extract aligned affordance knowledge from multiple images and integrate it into an affordance dictionary. Then, the Affordance Dictionary Adaptive Fusion Module (\textbf{ADM}) learns comprehensive point cloud representations that consider all affordance candidates in multiple images. Besides, the Multi-Image and Point Affordance (\textbf{MIPA}) benchmark is constructed and our method outperforms existing state-of-the-art methods on various experimental comparisons. Project page: \url{https://goxq.github.io/mifag}

Bin Zhao, Pengfei Han, Xuelong Li

Satellites are capable of capturing high-resolution videos. It makes vehicle perception from satellite become possible. Compared to street surveillance, drive recorder or other equipments, satellite videos provide a much broader city-scale view, so that the global dynamic scene of the traffic are captured and displayed. Traffic monitoring from satellite is a new task with great potential applications, including traffic jams prediction, path planning, vehicle dispatching, \emph{etc.}. Practically, limited by the resolution and view, the captured vehicles are very tiny (a few pixels) and move slowly. Worse still, these satellites are in Low Earth Orbit (LEO) to capture such high-resolution videos, so the background is also moving. Under this circumstance, traffic monitoring from the satellite view is an extremely challenging task. To attract more researchers into this field, we build a large-scale benchmark for traffic monitoring from satellite. It supports several tasks, including tiny object detection, counting and density estimation. The dataset is constructed based on 12 satellite videos and 14 synthetic videos recorded from GTA-V. They are separated into 408 video clips, which contain 7,336 real satellite images and 1,960 synthetic images. 128,801 vehicles are annotated totally, and the number of vehicles in each image varies from 0 to 101. Several classic and state-of-the-art approaches in traditional computer vision are evaluated on the datasets, so as to compare the performance of different approaches, analyze the challenges in this task, and discuss the future prospects. The dataset is available at: https://github.com/Chenxi1510/Vehicle-Perception-from-Satellite-Videos.

Chenjia Bai, Lingxiao Wang, Jianye Hao et al.

Offline Reinforcement Learning (RL) has shown promising results in learning a task-specific policy from a fixed dataset. However, successful offline RL often relies heavily on the coverage and quality of the given dataset. In scenarios where the dataset for a specific task is limited, a natural approach is to improve offline RL with datasets from other tasks, namely, to conduct Multi-Task Data Sharing (MTDS). Nevertheless, directly sharing datasets from other tasks exacerbates the distribution shift in offline RL. In this paper, we propose an uncertainty-based MTDS approach that shares the entire dataset without data selection. Given ensemble-based uncertainty quantification, we perform pessimistic value iteration on the shared offline dataset, which provides a unified framework for single- and multi-task offline RL. We further provide theoretical analysis, which shows that the optimality gap of our method is only related to the expected data coverage of the shared dataset, thus resolving the distribution shift issue in data sharing. Empirically, we release an MTDS benchmark and collect datasets from three challenging domains. The experimental results show our algorithm outperforms the previous state-of-the-art methods in challenging MTDS problems. See https://github.com/Baichenjia/UTDS for the datasets and code.

Bin Zhao, Xuelong Li, Xiaoqiang Lu et al.

Weather Recognition plays an important role in our daily lives and many computer vision applications. However, recognizing the weather conditions from a single image remains challenging and has not been studied thoroughly. Generally, most previous works treat weather recognition as a single-label classification task, namely, determining whether an image belongs to a specific weather class or not. This treatment is not always appropriate, since more than one weather conditions may appear simultaneously in a single image. To address this problem, we make the first attempt to view weather recognition as a multi-label classification task, i.e., assigning an image more than one labels according to the displayed weather conditions. Specifically, a CNN-RNN based multi-label classification approach is proposed in this paper. The convolutional neural network (CNN) is extended with a channel-wise attention model to extract the most correlated visual features. The Recurrent Neural Network (RNN) further processes the features and excavates the dependencies among weather classes. Finally, the weather labels are predicted step by step. Besides, we construct two datasets for the weather recognition task and explore the relationships among different weather conditions. Experimental results demonstrate the superiority and effectiveness of the proposed approach. The new constructed datasets will be available at https://github.com/wzgwzg/Multi-Label-Weather-Recognition.

Pingrui Zhang, Xianqiang Gao, Yuhan Wu et al.

In mobile manipulation, navigation and manipulation are often treated as separate problems, resulting in a significant gap between merely approaching an object and engaging with it effectively. Many navigation approaches primarily define success by proximity to the target, often overlooking the necessity for optimal positioning that facilitates subsequent manipulation. To address this, we introduce MoMa-Kitchen, a benchmark dataset comprising over 100k samples that provide training data for models to learn optimal final navigation positions for seamless transition to manipulation. Our dataset includes affordance-grounded floor labels collected from diverse kitchen environments, in which robotic mobile manipulators of different models attempt to grasp target objects amidst clutter. Using a fully automated pipeline, we simulate diverse real-world scenarios and generate affordance labels for optimal manipulation positions. Visual data are collected from RGB-D inputs captured by a first-person view camera mounted on the robotic arm, ensuring consistency in viewpoint during data collection. We also develop a lightweight baseline model, NavAff, for navigation affordance grounding that demonstrates promising performance on the MoMa-Kitchen benchmark. Our approach enables models to learn affordance-based final positioning that accommodates different arm types and platform heights, thereby paving the way for more robust and generalizable integration of navigation and manipulation in embodied AI. Project page: \href{https://momakitchen.github.io/}{https://momakitchen.github.io/}.

Kang Xu, Chenjia Bai, Xiaoteng Ma et al.

Generalizing policies across different domains with dynamics mismatch poses a significant challenge in reinforcement learning. For example, a robot learns the policy in a simulator, but when it is deployed in the real world, the dynamics of the environment may be different. Given the source and target domain with dynamics mismatch, we consider the online dynamics adaptation problem, in which case the agent can access sufficient source domain data while online interactions with the target domain are limited. Existing research has attempted to solve the problem from the dynamics discrepancy perspective. In this work, we reveal the limitations of these methods and explore the problem from the value difference perspective via a novel insight on the value consistency across domains. Specifically, we present the Value-Guided Data Filtering (VGDF) algorithm, which selectively shares transitions from the source domain based on the proximity of paired value targets across the two domains. Empirical results on various environments with kinematic and morphology shifts demonstrate that our method achieves superior performance compared to prior approaches.

Bin Zhao, Maoguo Gong, Xuelong Li

Although video summarization has achieved tremendous success benefiting from Recurrent Neural Networks (RNN), RNN-based methods neglect the global dependencies and multi-hop relationships among video frames, which limits the performance. Transformer is an effective model to deal with this problem, and surpasses RNN-based methods in several sequence modeling tasks, such as machine translation, video captioning, \emph{etc}. Motivated by the great success of transformer and the natural structure of video (frame-shot-video), a hierarchical transformer is developed for video summarization, which can capture the dependencies among frame and shots, and summarize the video by exploiting the scene information formed by shots. Furthermore, we argue that both the audio and visual information are essential for the video summarization task. To integrate the two kinds of information, they are encoded in a two-stream scheme, and a multimodal fusion mechanism is developed based on the hierarchical transformer. In this paper, the proposed method is denoted as Hierarchical Multimodal Transformer (HMT). Practically, extensive experiments show that HMT surpasses most of the traditional, RNN-based and attention-based video summarization methods.

Hailong Ning, Bin Zhao, Zhanxuan Hu et al.

The Dynamic Saliency Prediction (DSP) task simulates the human selective attention mechanism to perceive the dynamic scene, which is significant and imperative in many vision tasks. Most of existing methods only consider visual cues, while neglect the accompanied audio information, which can provide complementary information for the scene understanding. In fact, there exists a strong relation between auditory and visual cues, and humans generally perceive the surrounding scene by collaboratively sensing these cues. Motivated by this, an audio-visual collaborative representation learning method is proposed for the DSP task, which explores the audio modality to better predict the dynamic saliency map by assisting vision modality. The proposed method consists of three parts: 1) audio-visual encoding, 2) audio-visual location, and 3) collaborative integration parts. Firstly, a refined SoundNet architecture is adopted to encode audio modality for obtaining corresponding features, and a modified 3D ResNet-50 architecture is employed to learn visual features, containing both spatial location and temporal motion information. Secondly, an audio-visual location part is devised to locate the sound source in the visual scene by learning the correspondence between audio-visual information. Thirdly, a collaborative integration part is devised to adaptively aggregate audio-visual information and center-bias prior to generate the final saliency map. Extensive experiments are conducted on six challenging audiovisual eye-tracking datasets, including DIEM, AVAD, Coutrot1, Coutrot2, SumMe, and ETMD, which shows significant superiority over state-of-the-art DSP models.

Haopeng Li, Lingbo Liu, Kunlin Yang et al.

Video crowd localization is a crucial yet challenging task, which aims to estimate exact locations of human heads in the given crowded videos. To model spatial-temporal dependencies of human mobility, we propose a multi-focus Gaussian neighborhood attention (GNA), which can effectively exploit long-range correspondences while maintaining the spatial topological structure of the input videos. In particular, our GNA can also capture the scale variation of human heads well using the equipped multi-focus mechanism. Based on the multi-focus GNA, we develop a unified neural network called GNANet to accurately locate head centers in video clips by fully aggregating spatial-temporal information via a scene modeling module and a context cross-attention module. Moreover, to facilitate future researches in this field, we introduce a large-scale crowd video benchmark named VSCrowd, which consists of 60K+ frames captured in various surveillance scenarios and 2M+ head annotations. Finally, we conduct extensive experiments on three datasets including our SenseCrowd, and the experiment results show that the proposed method is capable to achieve state-of-the-art performance for both video crowd localization and counting.

Tianyi Zhang, Jie Lin, Peng Hu et al.

Non-maximum Suppression (NMS) is an essential postprocessing step in modern convolutional neural networks for object detection. Unlike convolutions which are inherently parallel, the de-facto standard for NMS, namely GreedyNMS, cannot be easily parallelized and thus could be the performance bottleneck in convolutional object detection pipelines. MaxpoolNMS is introduced as a parallelizable alternative to GreedyNMS, which in turn enables faster speed than GreedyNMS at comparable accuracy. However, MaxpoolNMS is only capable of replacing the GreedyNMS at the first stage of two-stage detectors like Faster-RCNN. There is a significant drop in accuracy when applying MaxpoolNMS at the final detection stage, due to the fact that MaxpoolNMS fails to approximate GreedyNMS precisely in terms of bounding box selection. In this paper, we propose a general, parallelizable and configurable approach PSRR-MaxpoolNMS, to completely replace GreedyNMS at all stages in all detectors. By introducing a simple Relationship Recovery module and a Pyramid Shifted MaxpoolNMS module, our PSRR-MaxpoolNMS is able to approximate GreedyNMS more precisely than MaxpoolNMS. Comprehensive experiments show that our approach outperforms MaxpoolNMS by a large margin, and it is proven faster than GreedyNMS with comparable accuracy. For the first time, PSRR-MaxpoolNMS provides a fully parallelizable solution for customized hardware design, which can be reused for accelerating NMS everywhere.

Bin Zhao, Xuelong Li

Video frame interpolation can up-convert the frame rate and enhance the video quality. In recent years, although the interpolation performance has achieved great success, image blur usually occurs at the object boundaries owing to the large motion. It has been a long-standing problem, and has not been addressed yet. In this paper, we propose to reduce the image blur and get the clear shape of objects by preserving the edges in the interpolated frames. To this end, the proposed Edge-Aware Network (EA-Net) integrates the edge information into the frame interpolation task. It follows an end-to-end architecture and can be separated into two stages, \emph{i.e.}, edge-guided flow estimation and edge-protected frame synthesis. Specifically, in the flow estimation stage, three edge-aware mechanisms are developed to emphasize the frame edges in estimating flow maps, so that the edge-maps are taken as the auxiliary information to provide more guidance to boost the flow accuracy. In the frame synthesis stage, the flow refinement module is designed to refine the flow map, and the attention module is carried out to adaptively focus on the bidirectional flow maps when synthesizing the intermediate frames. Furthermore, the frame and edge discriminators are adopted to conduct the adversarial training strategy, so as to enhance the reality and clarity of synthesized frames. Experiments on three benchmarks, including Vimeo90k, UCF101 for single-frame interpolation and Adobe240-fps for multi-frame interpolation, have demonstrated the superiority of the proposed EA-Net for the video frame interpolation task.

Bin Zhao, Maoguo Gong, Xuelong Li

Audio and vision are two main modalities in video data. Multimodal learning, especially for audiovisual learning, has drawn considerable attention recently, which can boost the performance of various computer vision tasks. However, in video summarization, existing approaches just exploit the visual information while neglect the audio information. In this paper, we argue that the audio modality can assist vision modality to better understand the video content and structure, and further benefit the summarization process. Motivated by this, we propose to jointly exploit the audio and visual information for the video summarization task, and develop an AudioVisual Recurrent Network (AVRN) to achieve this. Specifically, the proposed AVRN can be separated into three parts: 1) the two-stream LSTM is utilized to encode the audio and visual feature sequentially by capturing their temporal dependency. 2) the audiovisual fusion LSTM is employed to fuse the two modalities by exploring the latent consistency between them. 3) the self-attention video encoder is adopted to capture the global dependency in the video. Finally, the fused audiovisual information, and the integrated temporal and global dependencies are jointly used to predict the video summary. Practically, the experimental results on the two benchmarks, \emph{i.e.,} SumMe and TVsum, have demonstrated the effectiveness of each part, and the superiority of AVRN compared to those approaches just exploiting visual information for video summarization.

Bin Zhao, Haopeng Li, Xiaoqiang Lu et al.

Exploiting the inner-shot and inter-shot dependencies is essential for key-shot based video summarization. Current approaches mainly devote to modeling the video as a frame sequence by recurrent neural networks. However, one potential limitation of the sequence models is that they focus on capturing local neighborhood dependencies while the high-order dependencies in long distance are not fully exploited. In general, the frames in each shot record a certain activity and vary smoothly over time, but the multi-hop relationships occur frequently among shots. In this case, both the local and global dependencies are important for understanding the video content. Motivated by this point, we propose a Reconstructive Sequence-Graph Network (RSGN) to encode the frames and shots as sequence and graph hierarchically, where the frame-level dependencies are encoded by Long Short-Term Memory (LSTM), and the shot-level dependencies are captured by the Graph Convolutional Network (GCN). Then, the videos are summarized by exploiting both the local and global dependencies among shots. Besides, a reconstructor is developed to reward the summary generator, so that the generator can be optimized in an unsupervised manner, which can avert the lack of annotated data in video summarization. Furthermore, under the guidance of reconstruction loss, the predicted summary can better preserve the main video content and shot-level dependencies. Practically, the experimental results on three popular datasets i.e., SumMe, TVsum and VTW) have demonstrated the superiority of our proposed approach to the summarization task.

Xuelong Li, Kai Kou, Bin Zhao

In this paper, a new task is proposed, namely, weather translation, which refers to transferring weather conditions of the image from one category to another. It is important for photographic style transfer. Although lots of approaches have been proposed in traditional image translation tasks, few of them can handle the multi-category weather translation task, since weather conditions have rich categories and highly complex semantic structures. To address this problem, we develop a multi-domain weather translation approach based on generative adversarial networks (GAN), denoted as Weather GAN, which can achieve the transferring of weather conditions among sunny, cloudy, foggy, rainy and snowy. Specifically, the weather conditions in the image are determined by various weather-cues, such as cloud, blue sky, wet ground, etc. Therefore, it is essential for weather translation to focus the main attention on weather-cues. To this end, the generator of Weather GAN is composed of an initial translation module, an attention module and a weather-cue segmentation module. The initial translation module performs global translation during generation procedure. The weather-cue segmentation module identifies the structure and exact distribution of weather-cues. The attention module learns to focus on the interesting areas of the image while keeping other areas unaltered. The final generated result is synthesized by these three parts. This approach suppresses the distortion and deformation caused by weather translation. our approach outperforms the state-of-the-arts has been shown by a large number of experiments and evaluations.

Hailong Ning, Bin Zhao, Yuan Yuan

With the development of earth observation technology, massive amounts of remote sensing (RS) images are acquired. To find useful information from these images, cross-modal RS image-voice retrieval provides a new insight. This paper aims to study the task of RS image-voice retrieval so as to search effective information from massive amounts of RS data. Existing methods for RS image-voice retrieval rely primarily on the pairwise relationship to narrow the heterogeneous semantic gap between images and voices. However, apart from the pairwise relationship included in the datasets, the intra-modality and non-paired inter-modality relationships should also be taken into account simultaneously, since the semantic consistency among non-paired representations plays an important role in the RS image-voice retrieval task. Inspired by this, a semantics-consistent representation learning (SCRL) method is proposed for RS image-voice retrieval. The main novelty is that the proposed method takes the pairwise, intra-modality, and non-paired inter-modality relationships into account simultaneously, thereby improving the semantic consistency of the learned representations for the RS image-voice retrieval. The proposed SCRL method consists of two main steps: 1) semantics encoding and 2) semantics-consistent representation learning. Firstly, an image encoding network is adopted to extract high-level image features with a transfer learning strategy, and a voice encoding network with dilated convolution is devised to obtain high-level voice features. Secondly, a consistent representation space is conducted by modeling the three kinds of relationships to narrow the heterogeneous semantic gap and learn semantics-consistent representations across two modalities. Extensive experimental results on three challenging RS image-voice datasets show the effectiveness of the proposed method.

Bin Zhao, Xuelong Li, Xiaoqiang Lu

Exploiting the temporal dependency among video frames or subshots is very important for the task of video summarization. Practically, RNN is good at temporal dependency modeling, and has achieved overwhelming performance in many video-based tasks, such as video captioning and classification. However, RNN is not capable enough to handle the video summarization task, since traditional RNNs, including LSTM, can only deal with short videos, while the videos in the summarization task are usually in longer duration. To address this problem, we propose a hierarchical recurrent neural network for video summarization, called H-RNN in this paper. Specifically, it has two layers, where the first layer is utilized to encode short video subshots cut from the original video, and the final hidden state of each subshot is input to the second layer for calculating its confidence to be a key subshot. Compared to traditional RNNs, H-RNN is more suitable to video summarization, since it can exploit long temporal dependency among frames, meanwhile, the computation operations are significantly lessened. The results on two popular datasets, including the Combined dataset and VTW dataset, have demonstrated that the proposed H-RNN outperforms the state-of-the-arts.

Xuelong Li, Bin Zhao, Xiaoqiang Lu

In this paper, we build a general summarization framework for both of edited video and raw video summarization. Overall, our work can be divided into three folds: 1) Four models are designed to capture the properties of video summaries, i.e., containing important people and objects (importance), representative to the video content (representativeness), no similar key-shots (diversity) and smoothness of the storyline (storyness). Specifically, these models are applicable to both edited videos and raw videos. 2) A comprehensive score function is built with the weighted combination of the aforementioned four models. Note that the weights of the four models in the score function, denoted as property-weight, are learned in a supervised manner. Besides, the property-weights are learned for edited videos and raw videos, respectively. 3) The training set is constructed with both edited videos and raw videos in order to make up the lack of training data. Particularly, each training video is equipped with a pair of mixing-coefficients which can reduce the structure mess in the training set caused by the rough mixture. We test our framework on three datasets, including edited videos, short raw videos and long raw videos. Experimental results have verified the effectiveness of the proposed framework.

Xue Geng, Jie Fu, Bin Zhao et al.

This paper addresses a challenging problem - how to reduce energy consumption without incurring performance drop when deploying deep neural networks (DNNs) at the inference stage. In order to alleviate the computation and storage burdens, we propose a novel dataflow-based joint quantization approach with the hypothesis that a fewer number of quantization operations would incur less information loss and thus improve the final performance. It first introduces a quantization scheme with efficient bit-shifting and rounding operations to represent network parameters and activations in low precision. Then it restructures the network architectures to form unified modules for optimization on the quantized model. Extensive experiments on ImageNet and KITTI validate the effectiveness of our model, demonstrating that state-of-the-art results for various tasks can be achieved by this quantized model. Besides, we designed and synthesized an RTL model to measure the hardware costs among various quantization methods. For each quantization operation, it reduces area cost by about 15 times and energy consumption by about 9 times, compared to a strong baseline.