IterDet: Iterative Scheme for Object Detection in Crowded Environments

Rukhovich, Danila; Sofiiuk, Konstantin; Galeev, Danil; Barinova, Olga; Konushin, Anton

doi:10.48550/arxiv.2005.05708

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to enable the network with the ability to detect difficult pedestrians in challenging scenes, we train the model only with detection annotations in an end-toend way. During the training process, inspired by IterDet [39], we randomly split the ground truth detection bounding box set B gt of frame t into two subsets B his and B redis with B his ∪ B redis = B gt and B his ∩ B redis = ∅. We consider the aligned boxes set as B his in the first stage of data association and employ it to construct the history-aware mask H t .…”

Section: Network Architecturementioning

confidence: 99%

Improving Multiple Pedestrian Tracking in Crowded Scenes with Hierarchical Association

Xiao

Luo

2023

Entropy

View full text Add to dashboard Cite

Recently, advances in detection and re-identification techniques have significantly boosted tracking-by-detection-based multi-pedestrian tracking (MPT) methods and made MPT a great success in most easy scenes. Several very recent works point out that the two-step scheme of first detection and then tracking is problematic and propose using the bounding box regression head of an object detector to realize data association. In this tracking-by-regression paradigm, the regressor directly predicts each pedestrian’s location in the current frame according to its previous position. However, when the scene is crowded and pedestrians are close to each other, the small and partially occluded targets are easily missed. In this paper, we follow this pattern and design a hierarchical association strategy to obtain better performance in crowded scenes. To be specific, at the first association, the regressor is used to estimate the positions of obvious pedestrians. At the second association, we employ a history-aware mask to filter out the already occupied regions implicitly and look carefully at the remaining regions to find out the ignored pedestrians during the first association. We integrate the hierarchical association in a learning framework and directly infer the occluded and small pedestrians in an end-to-end way. We conduct extensive pedestrian tracking experiments on three public pedestrian tracking benchmarks from less crowded to crowded scenes, demonstrating the proposed strategy’s effectiveness in crowded scenes.

show abstract

Section: Network Architecturementioning

confidence: 99%