English

Depth as Prior Knowledge for Object Detection

Computer Vision and Pattern Recognition 2026-02-06 v1

Abstract

Detecting small and distant objects remains challenging for object detectors due to scale variation, low resolution, and background clutter. Safety-critical applications require reliable detection of these objects for safe planning. Depth information can improve detection, but existing approaches require complex, model-specific architectural modifications. We provide a theoretical analysis followed by an empirical investigation of the depth-detection relationship. Together, they explain how depth causes systematic performance degradation and why depth-informed supervision mitigates it. We introduce DepthPrior, a framework that uses depth as prior knowledge rather than as a fused feature, providing comparable benefits without modifying detector architectures. DepthPrior consists of Depth-Based Loss Weighting (DLW) and Depth-Based Loss Stratification (DLS) during training, and Depth-Aware Confidence Thresholding (DCT) during inference. The only overhead is the initial cost of depth estimation. Experiments across four benchmarks (KITTI, MS COCO, VisDrone, SUN RGB-D) and two detectors (YOLOv11, EfficientDet) demonstrate the effectiveness of DepthPrior, achieving up to +9% mAPS_S and +7% mARS_S for small objects, with inference recovery rates as high as 95:1 (true vs. false detections). DepthPrior offers these benefits without additional sensors, architectural changes, or performance costs. Code is available at https://github.com/mos-ks/DepthPrior.

Keywords

Cite

@article{arxiv.2602.05730,
  title  = {Depth as Prior Knowledge for Object Detection},
  author = {Moussa Kassem Sbeyti and Nadja Klein},
  journal= {arXiv preprint arXiv:2602.05730},
  year   = {2026}
}

Comments

This work has been submitted to the IEEE for possible publication

R2 v1 2026-07-01T09:38:02.487Z