Papers
1
Total Citations
4
H-Index
1
About
Shijie Ma is a researcher advancing the field of computer vision, with a particular focus on weakly-supervised learning and foundation models. Their most notable work, "WPS-SAM: Towards Weakly-Supervised Part Segmentation with Foundation Models" (2024), has already garnered 4 citations, signaling early impact in the community. This paper introduces a novel framework that leverages the Segment Anything Model (SAM) to achieve part-level segmentation with minimal annotation, addressing a critical bottleneck in fine-grained visual understanding. By combining the power of foundation models with weak supervision, Ma’s approach reduces the reliance on expensive pixel-level labels, making part segmentation more accessible for real-world applications like robotics and autonomous systems. Their contributions lie at the intersection of efficient learning and scalable vision, offering a practical pathway to bridge the gap between generic segmentation and detailed part analysis. As a rising voice in this domain, Shijie Ma’s work exemplifies how foundation models can be adapted for specialized tasks, promising to inspire further research in weakly-supervised techniques and their deployment in resource-constrained settings.
Research Focus
Key Achievements
Top Papers
- 1