BBRefinement: an universal scheme to improve precision of box object detectors

Petr Hurtik, Marek Vajgl

Keywords: object detection, deep neural networks, refinement

Abstract: We present a conceptually simple yet powerful and flexible scheme for refining predictions of bounding boxes. Our approach is trained standalone on GT boxes and can then be combined with an object detector to improve its predictions. The method, called BBRefinement, uses mixture data of image information and the object's class and center. Due to the transformation of the problem into a domain where BBRefinement does not care about multiscale detection, recognition of the object's class, computing confidence, or multiple detections, the training is much more effective. It results in the ability to refine even COCO's ground truth labels into a more precise form. BBRefinement improves the performance of SOTA architectures up to 2mAP points on the COCO dataset in the benchmark. The refinement process is fast; it adds 50-80ms overhead to a standard detector using RTX2080, so it can run in real-time on standard hardware. The code is available at https://gitlab.com/irafm-ai/bb-refinement.

One-sentence Summary: BBRefinement is a universal upgrade for existing object detectors to improve their accuracy via refinement of the detected bounding boxes.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=ZQ8ChTsxgy