Learning From Noisy Anchors for One-Stage Object Detection

By Hengduo Li, Zuxuan Wu, Chen Zhu, Caiming Xiong, Richard Socher, Larry S. Davis. 2019

112 citations recorded by the source corpus.

State-of-the-art object detectors rely on regressing and classifying an extensive list of possible anchors, which are divided into positive and negative samples based on their intersection-over-union (IoU) with corresponding ground-truth objects. Such a harsh split conditioned on IoU results in binary labels that are potentially noisy and challenging for training. In this paper, we propose to mitigate noise incurred by imperfect label assignment such that the contributions of anchors are dynamically determined by a carefully constructed cleanliness score associated with each anchor. Exploring outputs from both regression and classification branches, the cleanliness scores, estimated without incurring any additional computational overhead, are used not only as soft labels to supervise the training of the classification branch but also sample re-weighting factors for improved localization and classification accuracy. We conduct extensive experiments on COCO, and demonstrate, among other things, the proposed approach steadily improves RetinaNet by ~2% with various backbones.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Dacheng Tao
3. Chunhua Shen
4. Yi Yang
5. Jian Yang
6. Xiaogang Wang
7. Yu Qiao
8. Alan L. Yuille

Citations in the corpus

• Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection (2020, methodology): “Following the common practices in the official codes [3, 26, 31, 15], we also utilize the quality scores to weight LB and LD during training.”
• Probabilistic Anchor Assignment with IoU Prediction for Object Detection (2020, methodology): “Methods with a fixed number of positive samples [16,20] can assign these anchors as positives, and the model might predict these anchors with high scores during inference.”
• Detecting tiny objects in aerial images: A normalized Wasserstein distance and a new benchmark (2022, unknown): “Accurate object detection in the label noise scenario is of great interest both in the computer vision and earth vision community (Li et al., 2020a, 2019a; Pelletier et al., 2017).”
• Unsupervised Person Re-Identification via Multi-Label Classification (2020, background): “(Li et al., 2020) propose a cleanliness score based on classification and regression outputs to serve as the soft label and reweighting factor for object detector training.”
• DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution (2020, unknown): “, [40, 49, 54, 60, 64,73,86,87], and multi-stage methods, e.”
• TOOD: Task-aligned One-stage Object Detection (2021, methodology): “TAL differs from previous methods [1, 8, 9, 12, 14, 31] in two aspects.”
• OTA: Optimal Transport Assignment for Object Detection (2021, background): “Other recent advances [48, 19, 51, 16] suggest that the predicted confidence scores of each anchor could be a proper indicator to design dynamic assigning strategies, i.”
• Dynamic Anchor Learning for Arbitrary-Oriented Object Detection (2020, background): “Several attempts (Jiang et al. 2018; Li et al. 2020) have been made in this respect.”