End-to-End Detection and Re-identification Integrated Net for Person Search
This work addresses person search in real-world video surveillance where bounding box annotations are unavailable, representing an incremental improvement over existing joint detection and re-id methods.
This paper tackles the problem of person search in video surveillance by proposing an end-to-end integrated network (I-Net) that combines pedestrian detection and re-identification, narrowing the gap between these tasks and achieving superior performance on datasets.
This paper proposes a pedestrian detection and re-identification (re-id) integration net (I-Net) in an end-to-end learning framework. The I-Net is used in real-world video surveillance scenarios, where the target person needs to be searched in the whole scene videos, while the annotations of pedestrian bounding boxes are unavailable. By comparing to the OIM which is a work for joint detection and re-id, we have three distinct contributions. First, we introduce a Siamese architecture of I-Net instead of 1 stream, such that a verification task can be implemented. Second, we propose a novel on-line pairing loss (OLP) and hard example priority softmax loss (HEP), such that only the hard negatives are posed much attention in loss computation. Third, an on-line dictionary for negative samples storage is designed in I-Net without recording the positive samples. We show our result on person search datasets, the gap between detection and re-identification is narrowed. The superior performance can be achieved.