Unreliable-to-Reliable Instance Translation for Semi-Supervised Pedestrian Detection
- 10 February 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 24 (15209210) , 728-739
Abstract
Generating realistic pedestrian instances in a semi-supervised setting is promising but challenging due to the limited labeled data. We propose an unreliable-to-reliable instance translation model (Un2Reliab) conditioned on unreliable instances which poorly align with pedestrians. Un2Reliab mainly consists of an encoder-decoder-like generative network and a discriminative network, which are jointly trained in a minimax game. We adopt regularization to ensure that the synthesized instances are semantically similar to the corresponding ground truth. Furthermore, to preserve the identities of persons, we propose another regularization to ensure that the synthesized instances associated with the same person should be consistent in appearance. As a result, Un2Reliab learns to restore the missing parts of the original instances. As a side benefit, the synthesized instances are brought into better alignment. Inclusion of the synthesized data improves both the diversity and quality of training data, which eventually leads to better generalization performance. Extensive experiments indicate that Un2Reliab is able to synthesize high-fidelity pedestrian instances and improve the previous state-of-the-art results on multiple semi-supervised pedestrian detection benchmarks.Keywords
Funding Information
- National Natural Science Foundation of China (62072189)
- Research Grants Council of the Hong Kong Special Administration Region (11201220)
- Natural Science Foundation of Guangdong Province (2020A1515010484)
This publication has 48 references indexed in Scilit:
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Fast R-CNN2015
- Learning scene-specific pedestrian detectors without real data2015
- Data-driven 3D Voxel Patterns for object category recognition2015
- Cognitive pedestrian detector: Adapting detector to specific scene by transferring attributesNeurocomputing, 2015
- Scene-Specific Pedestrian Detection for Static Video SurveillanceIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Transferring a generic pedestrian detector towards specific scenes2012
- Multiobjective evolutionary algorithms to identify highly autocorrelated areas: the case of spatial distribution in financially compromised farmsAnnals of Operations Research, 2011
- Cascaded Confidence Filtering for Improved Tracking-by-Detection2010
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009