Multi-clue Consistency Learning to Bridge Gaps Between General and Oriented Object in Semi-supervised Detection

TL;DR

This paper introduces a novel semi-supervised detection framework called Multi-clue Consistency Learning (MCL) that effectively addresses the unique challenges of oriented object detection in aerial images, achieving state-of-the-art results.

Contribution

The paper proposes specific strategies like Gaussian Center Assignment, Scale-aware Label Assignment, and Consistent Confidence Soft Label to bridge gaps between general and oriented object detection in semi-supervised learning.

Findings

Achieves state-of-the-art performance on DOTA benchmarks.

Effectively handles oriented objects with various scales and aspect ratios.

Reduces noise and improves pseudo-label quality in semi-supervised detection.

Abstract

While existing semi-supervised object detection (SSOD) methods perform well in general scenes, they encounter challenges in handling oriented objects in aerial images. We experimentally find three gaps between general and oriented object detection in semi-supervised learning: 1) Sampling inconsistency: the common center sampling is not suitable for oriented objects with larger aspect ratios when selecting positive labels from labeled data. 2) Assignment inconsistency: balancing the precision and localization quality of oriented pseudo-boxes poses greater challenges which introduces more noise when selecting positive labels from unlabeled data. 3) Confidence inconsistency: there exists more mismatch between the predicted classification and localization qualities when considering oriented objects, affecting the selection of pseudo-labels. Therefore, we propose a Multi-clue Consistency Learning (MCL) framework to bridge gaps between general and oriented objects in semi-supervised detection. Specifically, considering various shapes of rotated objects, the Gaussian Center Assignment is specially designed to select the pixel-level positive labels from labeled data. We then introduce the Scale-aware Label Assignment to select pixel-level pseudo-labels instead of unreliable pseudo-boxes, which is a divide-and-rule strategy suited for objects with various scales. The Consistent Confidence Soft Label is adopted to further boost the detector by maintaining the alignment of the predicted results. Comprehensive experiments on DOTA-v1.5 and DOTA-v1.0 benchmarks demonstrate that our proposed MCL can achieve state-of-the-art performance in the semi-supervised oriented object detection task.