Research map / Continual learning / Class-incremental
PseDet: Revisiting the Power of Pseudo Label in Incremental Object Detection
Qiuchen Wang, Zehui Chen, Chenhongyi Yang, Jiaming Liu, Zhenyu Li, Feng Zhao
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes PseDet, an incremental object detection method that revisits pseudo labels to replay previously learned knowledge directly rather than only via response-level distillation.
Problem
Incremental object detection must learn new classes without forgetting old ones; existing response-level knowledge distillation replays memory only implicitly from the teacher on new data, which the authors argue underuses the knowledge generated by the model.
Approach
Revisits pseudo labels generated by the old model as a direct source of replayed knowledge for incremental detection, instead of relying solely on indirect response-level distillation. Only the abstract is available, so details are limited.
Closest papers by meaning
- YOLO-IOD: Towards Real Time Incremental Object Detection (AAAI 2026 · similarity 0.90)
- Gradient Decomposition and Alignment for Incremental Object Detection (ICCV 2025 · similarity 0.89)
- Bridging Non Co-occurrence with Unlabeled In-the-wild Data for Incremental Object Detection (NeurIPS 2021 · similarity 0.88)
- Continual Detection Transformer for Incremental Object Detection (CVPR 2023 · similarity 0.88)
- Revisiting Generative Replay for Class Incremental Object Detection (CVPR 2025 · similarity 0.88)
- DCA: Dividing and Conquering Amnesia in Incremental Object Detection (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
