On Model Calibration for Long-Tailed Object Detection and Instance Segmentation
arXiv:2107.02170
Abstract
Vanilla models for object detection and instance segmentation suffer from the heavy bias toward detecting frequent objects in the long-tailed setting. Existing methods address this issue mostly during training, e.g., by re-sampling or re-weighting. In this paper, we investigate a largely overlooked approach -- post-processing calibration of confidence scores. We propose NorCal, Normalized Calibration for long-tailed object detection and instance segmentation, a simple and straightforward recipe that reweighs the predicted scores of each class by its training sample size. We show that separately handling the background class and normalizing the scores over classes for each proposal are keys to achieving superior performance. On the LVIS dataset, NorCal can effectively improve nearly all the baseline models not only on rare classes but also on common and frequent classes. Finally, we conduct extensive analysis and ablation studies to offer insights into various modeling choices and mechanisms of our approach. Our code is publicly available at https://github.com/tydpan/NorCal/.
Accepted to NeurIPS 2021
References in corpus (8)
- YOLOv4: Optimal Speed and Accuracy of Object Detection
- On Calibration of Modern Neural Networks
- MMDetection: Open MMLab Detection Toolbox and Benchmark
- Frustratingly Simple Few-Shot Object Detection
- Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration
- Identifying and Compensating for Feature Deviation in Imbalanced Deep Learning
- 1st Place Solution of LVIS Challenge 2020: A Good Box is not a Guarantee of a Good Mask
- Classification Calibration for Long-tail Instance Segmentation