Publications (7)
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…
Inspiring the Next Generation of Segment Anything Models: Comprehensively Evaluate SAM and SAM 2 with Diverse Prompts Towards Context-Dependent Concepts under Different Scenes
Xiaoqi Zhao, Youwei Pang, Shijie Chang +10
As large-scale foundation models trained on billions of image--mask pairs covering a vast diversity of scenes, objects, and contexts, SAM and its upgraded version, SAM~2, have sign…
Towards Automatic Power Battery Detection: New Challenge, Benchmark Dataset and Baseline
Xiaoqi Zhao, Youwei Pang, Zhenyu Chen +5
We conduct a comprehensive study on a new task named power battery detection (PBD), which aims to localize the dense cathode and anode plates endpoints from X-ray images to evaluat…
Open-Vocabulary Camouflaged Object Segmentation
Youwei Pang, Xiaoqi Zhao, Jiaming Zuo +2
Recently, the emergence of the large-scale vision-language model (VLM), such as CLIP, has opened the way towards open-world object perception. Many works have explored the utilizat…
ConceptSeg-R1: Segment Any Concept via Meta-Reinforcement Learning
Yuan Zhao, Youwei Pang, Jiaming Zuo +10
Recent progress in promptable segmentation has shifted visual perception from object-level localization toward concept-level understanding. However, the notion of a concept remains…
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…
Spider: A Unified Framework for Context-dependent Concept Segmentation
Xiaoqi Zhao, Youwei Pang, Wei Ji +4
Different from the context-independent (CI) concepts such as human, car, and airplane, context-dependent (CD) concepts require higher visual understanding ability, such as camoufla…