Zhang Peiyuan

Zhang Peiyuan

WHU -> SJTU

I am an incoming Ph.D. student at Shanghai Jiao Tong University, advised by Prof. Xue Yang and Prof. Junchi Yan.

Previously, I was an undergraduate student at Wuhan University, where I worked with Prof. Yansheng Li.

My research interests include Deep Learning and Computer Vision, with a focus on Unified Fundation Model, Multimodal Large Language Model and Object Detection.

🔥 News

  • 2025.02:  🎉🎉 One paper related to OBB (Point2RBox-v2) is accepted by CVPR！
  • 2025.05:  🎉🎉 One paper related to OBB (PointOBB-v3) is accepted by IJCV！
  • 2025.07:  🎉🎉 One paper related to OBB (PWOOD) is now available on arXiv, Feel free to check it out.
  • 2025.09:  🎉🎉 One paper related to VLM (RISEBench) is accepted by NeurIPS Datasets and Benchmarks Track oral (Top 0.35%)！

📝 Publications

🔶Vision-Language Model

NeurIPS 2025 Oral
sym

Envisioning Beyond the Pixels: Benchmarking Reasoning-Informed Visual Editing

Xiangyu Zhao^, Peiyuan Zhang^, Kexian Tang^, Xiaorong Zhu^, Hao Li, Wenhao Chai, Zicheng Zhang, Renqiu Xia, Guangtao Zhai, Junchi Yan, Hua Yang°, Xue Yang°, Haodong Duan°

🌐Project

💡Summary This paper proposes RISEBench, the first benchmark for reasoning-informed visual editing, covering four core reasoning tasks—Temporal, Causal, Spatial, and Logical—and introducing a comprehensive evaluation framework with three key dimensions: Instruction Reasoning, Appearance Consistency, and Visual Plausibility.

🔷Object Detection

CVPR 2025
sym

Point2RBox-v2: Rethinking Point-supervised Oriented Object Detection with Spatial Layout Among Instances

Yi Yu^, Botao Ren^, Peiyuan Zhang^, Mingxin Liu, Junwei Luo, Shaofeng Zhang, Feipeng Da, Junchi Yan, Xue Yang°

🌐Project

💡Summary This work rethinks point-supervised oriented object detection with the layout among instances. At the core are three principles: 1) Gaussian overlap loss. 2) Voronoi watershed loss. 3) Consistency loss. These principles lead to strong performance.
IJCV 2025
sym

Pointobb-v3: Expanding Performance Boundaries of Single Point-Supervised Oriented Object Detection

Peiyuan Zhang^, Junwei Luo^, Xue Yang^, Yi Yu, Qingyun Li, Yue Zhou, Xiaosong Jia, Xudong Lu, Jingdong Chen, Xiang Li, Junchi Yan, Yansheng Li°

🌐Project

💡Summary This work presents an extended conference version of PointOBB, which incorporates a novel Scale-Sensitive Feature Fusion (SSFF) module to improve the model's capability of perceiving object scales, and further proposes an end-to-end optimized framework.
arxiv 2025
sym

Partial Weakly-Supervised Oriented Object Detection

Mingxin Liu, Peiyuan Zhang, Yuan Liu, Wei Zhang, Yue Zhou, Ning Liao, Ziyang Gong, Junwei Luo, Zhirui Wang, Yi Yu, Xue Yang°

🌐Project

💡Summary This paper proposes PWOOD, a cost-effective framework for oriented object detection that uses partially weak and unlabeled data through orientation- and scale-aware learning, achieving competitive performance with much lower annotation cost.

🏅 Honors and Awards

  • 2025.11 “The Challenge Cup” National Undergraduate extracurricular academic scientific and technological works competition National First Prize

  • 2025.10 Lei Jun Scholarship of CS, Wuhan University (Top 1%)

  • 2025.09 First-class Scholarship of CS, Wuhan University (Top 5%)

  • 2025.08 National College Students Computer System Capability Competition (XiaomiCup) National First Prize

  • ……

🎓 Educations

  • 2022.09 - now, Wuhan University, School of Computer Science.

💬 Invited Talks

  • Not yet — but my GPU has heard plenty of my research talks.

🧑‍💻 Internships

  • 2023.12 - 2025.06, WHU SkyEarth
  • 2025.12 - now, Tencent YouTu Lab