WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition

TL;DR

WeakSAM leverages the Segment Anything Model to improve weakly-supervised object detection and segmentation by addressing pseudo-label issues and reducing reliance on prompts, achieving state-of-the-art results.

Contribution

The paper introduces WeakSAM, a novel approach that combines SAM with adaptive pseudo-label generation and RoI regularization for enhanced weakly-supervised recognition.

Findings

WeakSAM outperforms previous methods on WSOD and WSIS benchmarks.

Achieves average improvements of 7.4% and 8.5%.

Effectively addresses PGT incompleteness and noise issues.

Abstract

Weakly supervised visual recognition using inexact supervision is a critical yet challenging learning problem. It significantly reduces human labeling costs and traditionally relies on multi-instance learning and pseudo-labeling. This paper introduces WeakSAM and solves the weakly-supervised object detection (WSOD) and segmentation by utilizing the pre-learned world knowledge contained in a vision foundation model, i.e., the Segment Anything Model (SAM). WeakSAM addresses two critical limitations in traditional WSOD retraining, i.e., pseudo ground truth (PGT) incompleteness and noisy PGT instances, through adaptive PGT generation and Region of Interest (RoI) drop regularization. It also addresses the SAM's problems of requiring prompts and category unawareness for automatic object detection and segmentation. Our results indicate that WeakSAM significantly surpasses previous state-of-the-art methods in WSOD and WSIS benchmarks with large margins, i.e. average improvements of 7.4% and 8.5%, respectively. The code is available at \url{https://github.com/hustvl/WeakSAM}.