arXiv subjects

Haopeng Li

Publications and source records attributed to Haopeng Li.

25 records · Page 2Linked to original sources

arXiv2024

Multimodal Semantic Communication for Generative Audio-Driven Video Conferencing

This paper studies an efficient multimodal data communication scheme for video conferencing. In our considered system, a speaker gives a talk to the audiences, with talking head video and audio being transmitted. Since the speaker does not frequently change posture and high-fidelity transmission of audio (speech and music) is required, redundant visual video data exists and can be removed by generating the video from the audio. To this end, we propose a wave-to-video (Wav2Vid) system, an efficient video transmission framework that reduces transmitted data by generating talking head video from audio. In particular, full-duration audio and short-duration video data are synchronously transmitted through a wireless channel, with neural networks (NNs) extracting and encoding audio and video semantics. The receiver then combines the decoded audio and video data, as well as uses a generative adversarial network (GAN) based model to generate the lip movement videos of the speaker. Simulation results show that the proposed Wav2Vid system can reduce the amount of transmitted data by up to 83% while maintaining the perceptual quality of the generated conferencing video.

cs.MM↗

arXiv2024

Visual-Text Cross Alignment: Refining the Similarity Score in Vision-Language Models

It has recently been discovered that using a pre-trained vision-language model (VLM), e.g., CLIP, to align a whole query image with several finer text descriptions generated by a large language model can significantly enhance zero-shot performance. However, in this paper, we empirically find that the finer descriptions tend to align more effectively with local areas of the query image rather than the whole image, and then we theoretically validate this finding. Thus, we present a method called weighted visual-text cross alignment (WCA). This method begins with a localized visual prompting technique, designed to identify local visual areas within the query image. The local visual areas are then cross-aligned with the finer descriptions by creating a similarity matrix using the pre-trained VLM. To determine how well a query image aligns with each category, we develop a score function based on the weighted similarities in this matrix. Extensive experiments demonstrate that our method significantly improves zero-shot performance across various datasets, achieving results that are even comparable to few-shot learning methods.

cs.CV↗

arXiv2024

Video Semantic Communication with Major Object Extraction and Contextual Video Encoding

This paper studies an end-to-end video semantic communication system for massive communication. In the considered system, the transmitter must continuously send the video to the receiver to facilitate character reconstruction in immersive applications, such as interactive video conference. However, transmitting the original video information with substantial amounts of data poses a challenge to the limited wireless resources. To address this issue, we reduce the amount of data transmitted by making the transmitter extract and send the semantic information from the video, which refines the major object and the correlation of time and space in the video. Specifically, we first develop a video semantic communication system based on major object extraction (MOE) and contextual video encoding (CVE) to achieve efficient video transmission. Then, we design the MOE and CVE modules with convolutional neural network based motion estimation, contextual extraction and entropy coding. Simulation results show that compared to the traditional coding schemes, the proposed method can reduce the amount of transmitted data by up to 25% while increasing the peak signal-to-noise ratio (PSNR) of the reconstructed video by up to 14%.

cs.NI↗

arXiv2022

Video Crowd Localization with Multi-focus Gaussian Neighborhood Attention and a Large-Scale Benchmark

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.

cs.CV↗

arXiv2022

Doubly Deformable Aggregation of Covariance Matrices for Few-shot Segmentation

Training semantic segmentation models with few annotated samples has great potential in various real-world applications. For the few-shot segmentation task, the main challenge is how to accurately measure the semantic correspondence between the support and query samples with limited training data. To address this problem, we propose to aggregate the learnable covariance matrices with a deformable 4D Transformer to effectively predict the segmentation map. Specifically, in this work, we first devise a novel hard example mining mechanism to learn covariance kernels for the Gaussian process. The learned covariance kernel functions have great advantages over existing cosine similarity-based methods in correspondence measurement. Based on the learned covariance kernels, an efficient doubly deformable 4D Transformer module is designed to adaptively aggregate feature similarity maps into segmentation results. By combining these two designs, the proposed method can not only set new state-of-the-art performance on public benchmarks, but also converge extremely faster than existing methods. Experiments on three public datasets have demonstrated the effectiveness of our method.

cs.CV↗

arXiv2021

Reconstructive Sequence-Graph Network for Video Summarization

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.

cs.CV↗

arXiv2019

Object Detection and 3D Estimation via an FMCW Radar Using a Fully Convolutional Network

This paper considers object detection and 3D estimation using an FMCW radar. The state-of-the-art deep learning framework is employed instead of using traditional signal processing. In preparing the radar training data, the ground truth of an object orientation in 3D space is provided by conducting image analysis, of which the images are obtained through a coupled camera to the radar device. To ensure successful training of a fully convolutional network (FCN), we propose a normalization method, which is found to be essential to be applied to the radar signal before feeding into the neural network. The system after proper training is able to first detect the presence of an object in an environment. If it does, the system then further produces an estimation of its 3D position. Experimental results show that the proposed system can be successfully trained and employed for detecting a car and further estimating its 3D position in a noisy environment.

cs.CV↗