Head-Tail Cooperative Learning Network for Unbiased Scene Graph Generation

TL;DR

This paper introduces a Head-Tail Collaborative Learning network for scene graph generation that balances prediction accuracy for both head and tail predicates, addressing long-tail distribution biases.

Contribution

It proposes a model-agnostic HTCL framework with collaborative feature branches and a self-supervised contrastive learning approach to improve unbiased scene graph generation.

Findings

Achieves higher mean Recall on VG150, Open Images V6, and GQA200 datasets.

Maintains a good balance between head and tail predicate prediction.

Sets new state-of-the-art performance in unbiased scene graph generation.

Abstract

Scene Graph Generation (SGG) as a critical task in image understanding, facing the challenge of head-biased prediction caused by the long-tail distribution of predicates. However, current unbiased SGG methods can easily prioritize improving the prediction of tail predicates while ignoring the substantial sacrifice in the prediction of head predicates, leading to a shift from head bias to tail bias. To address this issue, we propose a model-agnostic Head-Tail Collaborative Learning (HTCL) network that includes head-prefer and tail-prefer feature representation branches that collaborate to achieve accurate recognition of both head and tail predicates. We also propose a self-supervised learning approach to enhance the prediction ability of the tail-prefer feature representation branch by constraining tail-prefer predicate features. Specifically, self-supervised learning converges head predicate features to their class centers while dispersing tail predicate features as much as possible through contrast learning and head center loss. We demonstrate the effectiveness of our HTCL by applying it to various SGG models on VG150, Open Images V6 and GQA200 datasets. The results show that our method achieves higher mean Recall with a minimal sacrifice in Recall and achieves a new state-of-the-art overall performance. Our code is available at https://github.com/wanglei0618/HTCL.