BoxMask: Revisiting Bounding Box Supervision for Video Object Detection

TL;DR

BoxMask introduces a class-aware pixel-level supervision method for video object detection, significantly improving accuracy by refining object representations beyond traditional instance-level features.

Contribution

The paper proposes BoxMask, a novel approach that leverages bounding box annotations as coarse masks to enhance pixel-level discriminative features in video object detection.

Findings

Consistent improvement across ImageNet VID and EPIC KITCHENS datasets.

Effective integration into various state-of-the-art detectors.

Significant boost in detection accuracy with minimal additional complexity.

Abstract

We present a new, simple yet effective approach to uplift video object detection. We observe that prior works operate on instance-level feature aggregation that imminently neglects the refined pixel-level representation, resulting in confusion among objects sharing similar appearance or motion characteristics. To address this limitation, we propose BoxMask, which effectively learns discriminative representations by incorporating class-aware pixel-level information. We simply consider bounding box-level annotations as a coarse mask for each object to supervise our method. The proposed module can be effortlessly integrated into any region-based detector to boost detection. Extensive experiments on ImageNet VID and EPIC KITCHENS datasets demonstrate consistent and significant improvement when we plug our BoxMask module into numerous recent state-of-the-art methods.