3D PersonVLAD: Learning Deep Global Representations for Video-based Person Re-identification
This addresses the problem of accurately identifying individuals across video sequences for surveillance and security applications, representing an incremental improvement over existing methods.
The paper tackles video-based person re-identification by introducing 3D PersonVLAD, a global representation that aggregates local 3D features across full-length videos to capture appearance and motion dynamics, achieving state-of-the-art results on benchmark datasets like MARS, iLIDS-VID, and PRID 2011.
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