DINF: Dynamic Instance Noise Filter for Occluded Pedestrian Detection

TL;DR

This paper introduces DINF, a dynamic noise filtering method for RCNN-based pedestrian detection that enhances performance by reducing occlusion-related noise and addressing class imbalance.

Contribution

The paper proposes a novel dynamic instance noise filter and an IoU-Focal factor to improve occluded pedestrian detection in RCNN-based models.

Findings

Achieves state-of-the-art results on CrowdHuman and CityPersons datasets.

Effectively reduces noise in occluded instance features.

Balances training focus on severely overlapping objects.

Abstract

Occlusion issue is the biggest challenge in pedestrian detection. RCNN-based detectors extract instance features by cropping rectangle regions of interest in the feature maps. However, the visible pixels of the occluded objects are limited, making the rectangle instance feature mixed with a lot of instance-irrelevant noise information. Besides, by counting the number of instances with different degrees of overlap of CrowdHuman dataset, we find that the number of severely overlapping objects and the number of slightly overlapping objects are unbalanced, which may exacerbate the challenges posed by occlusion issues. Regarding to the noise issue, from the perspective of denoising, an iterable dynamic instance noise filter (DINF) is proposed for the RCNN-based pedestrian detectors to improve the signal-noise ratio of the instance feature. Simulating the wavelet denoising process, we use the instance feature vector to generate dynamic convolutional kernels to transform the RoIs features to a domain in which the near-zero values represent the noise information. Then, soft thresholding with channel-wise adaptive thresholds is applied to convert the near-zero values to zero to filter out noise information. For the imbalance issue, we propose an IoU-Focal factor (IFF) to modulate the contributions of the well-regressed boxes and the bad-regressed boxes to the loss in the training process, paying more attention to the minority severely overlapping objects. Extensive experiments conducted on CrowdHuman and CityPersons demonstrate that our methods can help RCNN-based pedestrian detectors achieve state-of-the-art performance.