# 3D PersonVLAD: Learning Deep Global Representations for Video-based Person Re-identification

## TL;DR

This paper proposes a novel 3D PersonVLAD method that captures appearance and motion in full-length videos for person re-identification, outperforming existing 2D-based approaches by effectively aggregating local features with 3D part alignment.

## Contribution

Introduction of 3D PersonVLAD, a global video representation layer that captures full-sequence features and includes a 3D part alignment module for improved local feature aggregation.

## Key findings

- Achieved state-of-the-art results on MARS, iLIDS-VID, and PRID 2011 datasets.
- Demonstrated superiority over 2D ConvNet-based methods in capturing motion and appearance.
- Effectively handled occlusions and misalignments through 3D part alignment.

## Abstract

In this paper, we introduce a global video representation to video-based person re-identification (re-ID) that aggregates local 3D features across the entire video extent. Most of the existing methods rely on 2D convolutional networks (ConvNets) to extract frame-wise deep features which are pooled temporally to generate the video-level representations. However, 2D ConvNets lose temporal input information immediately after the convolution, and a separate temporal pooling is limited in capturing human motion in shorter sequences. To this end, we present a \textit{global} video representation (3D PersonVLAD), complementary to 3D ConvNets as a novel layer to capture the appearance and motion dynamics in full-length videos. However, encoding each video frame in its entirety and computing an aggregate global representation across all frames is tremendously challenging due to occlusions and misalignments. To resolve this, our proposed network is further augmented with 3D part alignment module to learn local features through soft-attention module. These attended features are statistically aggregated to yield identity-discriminative representations. Our global 3D features are demonstrated to achieve state-of-the-art results on three benchmark datasets: MARS \cite{MARS}, iLIDS-VID \cite{VideoRanking}, and PRID 2011