A Dual-Path Model With Adaptive Attention for Vehicle Re-Identification

Abstract

In recent years, attention models have been extensively used for person and vehicle re-identification. Most reidentification methods are designed to focus attention on key-point locations. However, depending on the orientation, the contribution of each key-point varies. In this paper, we present a novel dual-path adaptive attention model for vehicle re-identification (AAVER). The global appearance path captures macroscopic vehicle features while the orientation conditioned part appearance path learns to capture localized discriminative features by focusing attention on the most informative key-points. Through extensive experimentation, we show that the proposed AAVER method is able to accurately re-identify vehicles in unconstrained scenarios, yielding state of the art results on the challenging dataset VeRi-776. As a byproduct, the proposed system is also able to accurately predict vehicle key-points and shows an improvement of more than 7% over state of the art. The code for key-point estimation model is available at https://github.com/Pirazh/Vehicle_Key_Point_Orientation_Estimation

A Strong and Efficient Baseline for Vehicle Re-Identification Using Deep Triplet Embedding

“…Ours 128 ID GSTE [2] 1024 ID VAMI [62] 2048 ID + A OIFE [53] 256 ID + K MGR [58] 1024 ID + A ATT [58] 1024 ID + A C2F [9] 1024 ID + A CLVR [19] 1024 A PAMTRI (All)* [49] 1024 ID + K + A MSVR [20] 2048 ID FDA-Net [31] 1024 ID AAVER [21] 2048 ID + K Referring to the best results in Table 2, in the subsequent Sections we consider only triplet loss without embedding-normalization.…”

“…PAMTRI (RS) uses mix of real and synthetic data for learning embedding, while PAMTRI (All) additionally utilizes vehicle keypoints and attributes in a multi-task learning framework. AAVER [21] is a recent work presented at ICCV 2019. Authors construct a dual path network for extracting global and local features.…”

“…Contrary to our method, other approaches [9,28,58], all utilize model annotations (in addition to identity annotations) from the training set for re-identification. Similarly to OIFE [53], AAVER [21] uses identitiy and orientation information for training the embedding network. We consider the image-to-image search (or image based re-identification) subset of this dataset.…”

Multi-Proxy Constraint Loss for Vehicle Re-Identification

“…Tang et al [ 9 ] reasoned the vehicle pose and shape with synthetic datasets and passed this information to the attributes and feature learning network. Khorramshahi et al [ 10 ] increased a path to detect vehicle key points using the orientation as a conditional factor and extract the local features to distinguish similar vehicles. However, these multi-view approaches require additional labels of key points or viewpoints and complex training process.…”

“…Some work [ 6 , 7 , 8 , 9 , 10 , 11 ] were devoted to addressing the intra-class variance problem of vehicle re-identification by predicting key points or viewpoints. The key points can be passed as input to feature extract network [ 9 ] or directly used as the discriminant regions to aggregate the orientation-invariant features [ 6 ] and trained supervised by IDs to distinguish similar vehicles [ 10 ]. Despite obtaining local discriminative features, key points require extra labels and are only partially visible in different viewpoints.…”