Revisiting RCNN: On Awakening the Classification Power of Faster RCNN
Bowen Cheng, Yunchao Wei, Honghui Shi, Rogerio Feris, JinJun Xiong, Thomas Huang
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/MindSpore-paper-code-3/code10/tree/main/faster_rcnn_ssodmindspore★ 0
- github.com/MindSpore-paper-code-3/code3/tree/main/faster_rcnn_ssodmindspore★ 0
- github.com/MindSpore-paper-code-3/code8/tree/main/faster_rcnn_ssodmindspore★ 0
- github.com/MindSpore-paper-code-3/code7/tree/main/faster_rcnn_ssodmindspore★ 0
- github.com/makefile/DCRtf★ 0
- github.com/SHI-Labs/Decoupled-Classification-Refinementmxnet★ 0
- github.com/bowenc0221/Decoupled-Classification-Refinementtf★ 0
Abstract
Recent region-based object detectors are usually built with separate classification and localization branches on top of shared feature extraction networks. In this paper, we analyze failure cases of state-of-the-art detectors and observe that most hard false positives result from classification instead of localization. We conjecture that: (1) Shared feature representation is not optimal due to the mismatched goals of feature learning for classification and localization; (2) multi-task learning helps, yet optimization of the multi-task loss may result in sub-optimal for individual tasks; (3) large receptive field for different scales leads to redundant context information for small objects.We demonstrate the potential of detector classification power by a simple, effective, and widely-applicable Decoupled Classification Refinement (DCR) network. DCR samples hard false positives from the base classifier in Faster RCNN and trains a RCNN-styled strong classifier. Experiments show new state-of-the-art results on PASCAL VOC and COCO without any bells and whistles.