C
Can Peng
Researcher at University of Queensland
Publications - 11
Citations - 100
Can Peng is an academic researcher from University of Queensland. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 3, co-authored 8 publications receiving 24 citations.
Papers
More filters
Journal ArticleDOI
Faster ILOD: Incremental learning for object detectors based on faster RCNN
TL;DR: This paper designs an efficient end-to-end incremental object detector using knowledge distillation for object detectors based on RPNs and introduces multi-network adaptive distillation that properly retains knowledge from the old categories when fine-turning the model for new task.
Journal ArticleDOI
SID: Incremental learning for anchor-free object detection via Selective and Inter-related Distillation
TL;DR: A novel incremental learning paradigm called Selective and Inter-related Distillation (SID) is proposed and a novel evaluation metric is proposed to better assess the performance of detectors under incremental learning conditions.
Proceedings ArticleDOI
Few-Shot Class-Incremental Learning from an Open-Set Perspective
TL;DR: This paper reevaluate the current task setting and proposes a more comprehensive and practical setting for the FSCIL task, and proposes the method — Augmented Angular Loss Incremental Classification or ALICE, which uses the angular penalty loss to obtain well-clustered features.
Posted Content
Faster ILOD: Incremental Learning for Object Detectors based on Faster RCNN
TL;DR: In this article, the authors proposed an end-to-end incremental object detector using knowledge distillation, which uses a multi-network adaptive distillation that properly retains knowledge from the old categories when fine-tuning the model for new task.
Posted Content
To What Extent Does Downsampling, Compression, and Data Scarcity Impact Renal Image Analysis?
TL;DR: This paper shows that the image file size of 40 χ WSI images can be reduced by a factor of over 6000 with negligible loss of glomerulus detection accuracy, and examines the impact of image downsampling rates and compression on detection accuracy.