Object Permanence in Object Detection Leveraging Temporal Priors at Inference Time

TL;DR

This paper introduces a method to incorporate object permanence into two-stage object detectors by using temporal priors from previous frames at inference time, significantly improving detection stability under occlusion.

Contribution

The authors propose a novel inference-time approach inspired by particle filters that enhances two-stage detectors with explicit object permanence without retraining.

Findings

Detection performance improved by up to 10.3 mAP.

Method maintains low computational overhead.

Effective under heavy occlusion.

Abstract

Object permanence is the concept that objects do not suddenly disappear in the physical world. Humans understand this concept at young ages and know that another person is still there, even though it is temporarily occluded. Neural networks currently often struggle with this challenge. Thus, we introduce explicit object permanence into two stage detection approaches drawing inspiration from particle filters. At the core, our detector uses the predictions of previous frames as additional proposals for the current one at inference time. Experiments confirm the feedback loop improving detection performance by a up to 10.3 mAP with little computational overhead. Our approach is suited to extend two-stage detectors for stabilized and reliable detections even under heavy occlusion. Additionally, the ability to apply our method without retraining an existing model promises wide application in real-world tasks.