Vehicle Re-Identification Using Quadruple Directional Deep Learning Features

Abstract

In order to resist the adverse effect of viewpoint variations for improving vehicle re-identification performance, we design quadruple directional deep learning networks to extract quadruple directional deep learning features (QD-DLF) of vehicle images. The quadruple directional deep learning networks are with similar overall architecture, including the same basic deep learning architecture but different directional feature pooling layers. Specifically, the same basic deep learning architecture is a shortly and densely connected convolutional neural network to extract basic feature maps of an input square vehicle image in the first stage. Then, the quadruple directional deep learning networks utilize different directional pooling layers, i.e., horizontal average pooling (HAP) layer, vertical average pooling (VAP) layer, diagonal average pooling (DAP) layer and anti-diagonal average pooling (AAP) layer, to compress the basic feature maps into horizontal, vertical, diagonal and anti-diagonal directional feature maps, respectively. Finally, these directional feature maps are spatially normalized and concatenated together as a quadruple directional deep learning feature for vehicle re-identification. Extensive experiments on both VeRi and VehicleID databases show that the proposed QD-DLF approach outperforms multiple state-of-the-art vehicle re-identification methods. arXiv:1811.05163v1 [cs.CV] 13 Nov 2018Jianqing Zhu received the B.S. degree in communication engineering and the M.S. degree in communication and information system from the

Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification

“…First, we considered LOMO [39], which utilizes a handcrafted local feature for person re-ID; it solves the problems associated with view and illumination variations. The GoogLeNet fine-tuned on the CompCars dataset [40] can extract high-level semantic attributes of the vehicle appearance, while VAMI [41] is a viewpoint-aware attention model used to extract the core area from different views through an adversarial network, and QD-DLF [42] has different directional feature pooling layers. Siamese-CNN + Path-LSTM [18] is a two-stage framework that combines complex spatiotemporal information and effectively regularizes the re-ID results.…”

Robust, Extensible, and Fast: Teamed Classifiers for Vehicle Tracking in Multi-Camera Networks

“…Challenges lie in re-id under adversarial conditions where vehicles need to be tracked with multi-orientation, multi-scale, multi-resolution values alongside possible occlusion and blur. The vehicle reidentification problem has seen significant work in the past few years due to advances in the general one-shot learning problem [10], [15], [16], [17], [18], [19], [20]. • Event detection: Automated event detection remains a difficult challenge due to the lack of labeled real-world or synthetic data and absence of frameworks for video-based anomaly detection.…”

“…Vehicle re-identification. More recently, there have been approaches for end-to-end vehicle metadata extraction and re-identification in [10], [16], [17], [23], [19], [20]. OIFE [23] proposed stacked convolutional networks (SCN) to extract finegrained features in conjunction with global features.…”

“…Conversely, existing vehicle re-id datasets such as VeRi-776 [35] and VeRi-Wild [36] primarily focus on intraclass variability. Current approaches in vehicle re-id attempt to address inter-class similarity and intra-class variability in the same end-to-end model [16], [17], [18], [20]. This creates models that sacrifice performance on solving edge cases in intra-class variability to increase discriminative ability for interclass similarity across the entire data space.…”

“…We can build smaller models compared to recent approaches. As such, each member of the intra-class variability team uses a single ResNet 18 backbone and can operate in near real-time, compared to 20 stacked convolutional networks in [23], 5 ResNet backbones in [16], and 4 ResNet50 backbones in [20].…”