Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement

TL;DR

Salience DETR introduces hierarchical salience filtering and query refinement to improve detection accuracy and efficiency in transformer-based object detection, reducing computational load while enhancing performance.

Contribution

The paper proposes a novel hierarchical salience filtering refinement method for DETR, addressing scale bias and semantic misalignment issues in two-stage detection frameworks.

Findings

Achieves +4.0% AP improvement on three detection datasets

Attains 49.2% AP on COCO 2017 with fewer FLOPs

Demonstrates better trade-off between efficiency and accuracy

Abstract

DETR-like methods have significantly increased detection performance in an end-to-end manner. The mainstream two-stage frameworks of them perform dense self-attention and select a fraction of queries for sparse cross-attention, which is proven effective for improving performance but also introduces a heavy computational burden and high dependence on stable query selection. This paper demonstrates that suboptimal two-stage selection strategies result in scale bias and redundancy due to the mismatch between selected queries and objects in two-stage initialization. To address these issues, we propose hierarchical salience filtering refinement, which performs transformer encoding only on filtered discriminative queries, for a better trade-off between computational efficiency and precision. The filtering process overcomes scale bias through a novel scale-independent salience supervision. To compensate for the semantic misalignment among queries, we introduce elaborate query refinement modules for stable two-stage initialization. Based on above improvements, the proposed Salience DETR achieves significant improvements of +4.0% AP, +0.2% AP, +4.4% AP on three challenging task-specific detection datasets, as well as 49.2% AP on COCO 2017 with less FLOPs. The code is available at https://github.com/xiuqhou/Salience-DETR.