About me

We propose Collaborative Anatomy Diffusion (CA-Diff), a framework integrating spatial anatomical features to enhance segmentation accuracy of the diffusion model.

We propose a Self-Supervised Fragment FineTuning (SF2T), a novel effortless fine-tuning method, employs the rich inherent characteristics of videos for training, while unlocking more fine-grained understanding ability of Video-LLMs.

We introduce Ref-GS, a novel approach for directional light factorization in 2D Gaussian splatting, which enables photorealistic view-dependent appearance rendering and precise geometry recovery.

We propose DEEM, a simple but effective approach that utilizes the generative feedback of diffusion models to align the semantic distributions of the image encoder.

We propose a section-based multi-object tracking approach that integrates a temporal coherent Object Flow Tracker, capable of achieving simultaneous multi-frame tracking by treating multiple consecutive frames as the basic processing unit, denoted as a “section”.

We introduce innovative motion and appearance conditions that are seamlessly integrated into our patch diffusion model.

We propose the Coupled SSM model, for coupling state chains of multiple modalities while maintaining independence of intra-modality state processes.

We introduce a novel approach leveraging language embeddings to enhance the coherence of frame-wise visual features related to the same object.

We can apply compression operation to a deeper degree for less irrelevant parts to achieve a high compression rate, while preserving the performance by applying a lower compression ratio to the more important parts.

We propose the Adaptable Motion Diffusion (AMD) model, which leverages a Large Language Model (LLM) to parse the input text into a sequence of concise and interpretable anatomical scripts that correspond to the target motion.

We propose to harness the capabilities of a Large Language Model (LLM) to decompose text descriptions into coherent directives adhering to stringent formats and progressively generate the target image.

We formulates object detection and association jointly as a consistent denoising diffusion process from paired noise boxes to paired ground-truth boxes.

We demonstrate the basic vision transformer (ViT) architecture is sufficient for visual tracking with correlative masked modeling for information aggregation enhancement.

CSWinTT is a new transformer architecture with multi-scale cyclic shifting window attention for visual object tracking, elevating the attention from pixel to window level.

We proposed a distractor-aware player tracking algorithm and a high-quality benchmark for soccer play tracking, deal with occlusion and similar distractors in soccer scenes.

We designed and developed a sports video search engine based on distributed architecture, aimimng to provide content-based video analysis and retrieval services

We construct a soccer dataset named Soccer Dataset for Shot, Event, and Tracking, to meet the research needs of shot segmentation, event detection and player tracking

We focus on broadcast soccer videos and present a comprehensive dataset for analysis, including shot boundaries, event annotations, and bounding boxes.