106 citations · 113 across the 17 of their papers we have counts for
7 papers · 1 filter
Rethinking Evaluation of Infrared Small Target Detection
Youwei Pang, Xiaoqi Zhao, Lihe Zhang +4
As an essential vision task, infrared small target detection (IRSTD) has seen significant advancements through deep learning. However, critical limitations in current evaluation pr…
UniMRSeg: Unified Modality-Relax Segmentation via Hierarchical Self-Supervised Compensation
Xiaoqi Zhao, Youwei Pang, Chenyang Yu +5
Multi-modal image segmentation faces real-world deployment challenges from incomplete/corrupted modalities degrading performance. While existing methods address training-inference…
UniMMAD: Unified Multi-Modal and Multi-Class Anomaly Detection via MoE-Driven Feature Decompression
Yuan Zhao, Youwei Pang, Lihe Zhang +4
Existing anomaly detection (AD) methods often treat the modality and class as independent factors. Although this paradigm has enriched the development of AD research branches and p…
Referring Remote Sensing Image Segmentation with Cross-view Semantics Interaction Network
Jiaxing Yang, Lihe Zhang, Huchuan Lu
Recently, Referring Remote Sensing Image Segmentation (RRSIS) has aroused wide attention. To handle drastic scale variation of remote targets, existing methods only use the full im…
Power Battery Detection
Xiaoqi Zhao, Peiqian Cao, Chenyang Yu +10
Power batteries are essential components in electric vehicles, where internal structural defects can pose serious safety risks. We conduct a comprehensive study on a new task, powe…
UniSegDiff: Boosting Unified Lesion Segmentation via a Staged Diffusion Model
Yilong Hu, Shijie Chang, Lihe Zhang +3
The Diffusion Probabilistic Model (DPM) has demonstrated remarkable performance across a variety of generative tasks. The inherent randomness in diffusion models helps address issu…
EffOWT: Transfer Visual Language Models to Open-World Tracking Efficiently and Effectively
Bingyang Wang, Kaer Huang, Bin Li +4
Open-World Tracking (OWT) aims to track every object of any category, which requires the model to have strong generalization capabilities. Trackers can improve their generalization…