Partial Person Re-identification with Alignment and Hallucination
This addresses a practical challenge in CCTV surveillance where only partial views of pedestrians are available, though it is incremental as it builds on existing re-identification techniques.
The paper tackles the problem of partial person re-identification by proposing a method that aligns and hallucinates missing body parts, achieving significant improvements in matching accuracy across three datasets.
Partial person re-identification involves matching pedestrian frames where only a part of a body is visible in corresponding images. This reflects practical CCTV surveillance scenario, where full person views are often not available. Missing body parts make the comparison very challenging due to significant misalignment and varying scale of the views. We propose Partial Matching Net (PMN) that detects body joints, aligns partial views and hallucinates the missing parts based on the information present in the frame and a learned model of a person. The aligned and reconstructed views are then combined into a joint representation and used for matching images. We evaluate our approach and compare to other methods on three different datasets, demonstrating significant improvements.