Proposal Learning for Semi-Supervised Object Detection
Peng Tang† Chetan Ramaiah† Yan Wang‡ Ran Xu† Caiming Xiong† †Salesforce Research ‡The Johns Hopkins University
Abstract
In this paper, we focus on semi-supervised object detection to boost performance of proposal-based object detectors (a.k.a. two-stage object detectors) by training on both labeled and unlabeled data. However, it is non-trivial to train object detectors on unlabeled data due to the unavailability of ground truth labels. To address this problem, we present a proposal learning approach to learn proposal features and predictions from both labeled and unlabeled data. The approach consists of a self-supervised proposal learning module and a consistency-based proposal learning module. In the self-supervised proposal learning module, we present a proposal location loss and a contrastive loss to learn context-aware and noise-robust proposal features respectively. In the consistency-based proposal learning module, we apply consistency losses to both bounding box classification and regression predictions of proposals to learn noise-robust proposal features and predictions. Our approach enjoys the following benefits: 1) encouraging more context information to delivered in the proposals learning procedure; 2) noisy proposal features and enforcing consistency to allow noise-robust object detecti
中文速览
半监督目标检测（Semi-Supervised Object Detection）长期面临一个难题：大量无标注图片无法直接用于训练需要边界框标签的两阶段检测器。针对这一问题，研究者提出了一种"提案学习"（proposal learning）方法，专门针对 Faster R-CNN 这类先生成候选框、再做分类和定位的检测器，从有标注和无标注数据中同时学习候选框特征与预测结果。具体做法分两部分：一是自监督模块，用预测候选框位置的"借口任务"让模型感知更多上下文信息，同时用对比学习让特征对噪声保持鲁棒；二是一致性模块，给候选框特征加噪后要求分类和回归预测结果与原始特征保持一致，从而进一步增强抗噪能力。在 COCO 基准上，该方法相比纯监督基线平均提升约 2.0% AP，与数据蒸馏（data distillation）结合后还能再多涨约 0.9%，证明两者具有互补性。这项工作的意义在于，它提供了一个通用框架，无需额外标注成本就能持续提升两阶段检测器的性能，对实际中标注资源有限的场景极具参考价值。
原文 arXiv:2001.05086;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2001.05086v2