Generalization Limits in Vehicle Re-Identification

• URL: http://arxiv.org/abs/2606.01981v1
• Date: Mon, 01 Jun 2026 09:43:06 GMT
• Title: Generalization Limits in Vehicle Re-Identification
• Abstract summary: Vehicle re-identification focuses on retrieving images of the same vehicle from a gallery given a query image.<n>We observe that vehicles with few visual differences-e.g., the same make, model, and color-appear in both the training and test sets.<n>We propose a novel evaluation approach that more effectively measures generalization capability to unseen vehicle types.
• Score: 15.820955344313061
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vehicle re-identification focuses on retrieving images of the same vehicle from a gallery given a query image. Upon closer inspection of commonly used datasets, we observe that vehicles with few visual differences-e.g., the same make, model, and color-appear in both the training and test sets. As a result, methods that effectively memorize the training data tend to perform well on these test sets but struggle to generalize to other datasets. In this paper, we address this issue by proposing a novel evaluation approach that more effectively measures generalization capability to unseen vehicle types. To further study generalization performance, we also propose splitting the evaluation based on view, allowing us to differentiate the effect of viewpoint robustness from that of same-view re-identification. Our findings reveal that most state-of-the-art methods struggle with unseen vehicle types, and that their robustness to viewpoint changes and attention to detail are limited to vehicle types seen during training.

Related papers

• PoseDriver: A Unified Approach to Multi-Category Skeleton Detection for Autonomous Driving [72.70946892443236]
  PoseDriver is a unified framework for bottom-up multi-category skeleton detection tailored to common objects in driving scenarios.<n>We propose a novel approach for lane detection based on skeleton representations, achieving state-of-the-art performance on the OpenLane dataset.<n>We present a new dataset for bicycle skeleton detection and assess the transferability of our framework to novel categories.
  arXiv  Detail & Related papers  (2026-03-24T13:56:35Z)
• CLIPVehicle: A Unified Framework for Vision-based Vehicle Search [13.316099306091239]
  We propose a new unified framework, namely CLIPVehicle, which contains a dual-granularity semantic-region alignment module.<n>We also construct a new benchmark, including a real-world dataset CityFlowVS, and two synthetic datasets SynVS-Day and SynVS-All, for vehicle search.
  arXiv  Detail & Related papers  (2025-08-06T06:36:44Z)
• Image-Based Vehicle Classification by Synergizing Features from Supervised and Self-Supervised Learning Paradigms [0.913755431537592]
  Two state-of-the-art self-supervised learning methods, DINO and data2vec, were evaluated and compared for their representation learning of vehicle images. The representations learned from these self-supervised learning methods were combined with the wheel positional features for the vehicle classification task. Our experiments show that the data2vec-distilled representations, which are consistent with our wheel masking strategy, outperformed the DINO counterpart.
  arXiv  Detail & Related papers  (2023-02-01T18:22:23Z)
• Self-Supervised Steering Angle Prediction for Vehicle Control Using Visual Odometry [55.11913183006984]
  We show how a model can be trained to control a vehicle's trajectory using camera poses estimated through visual odometry methods. We propose a scalable framework that leverages trajectory information from several different runs using a camera setup placed at the front of a car.
  arXiv  Detail & Related papers  (2021-03-20T16:29:01Z)
• Pluggable Weakly-Supervised Cross-View Learning for Accurate Vehicle Re-Identification [53.6218051770131]
  Cross-view consistent feature representation is key for accurate vehicle ReID. Existing approaches resort to supervised cross-view learning using extensive extra viewpoints annotations. We present a pluggable Weakly-supervised Cross-View Learning (WCVL) module for vehicle ReID.
  arXiv  Detail & Related papers  (2021-03-09T11:51:09Z)
• Vehicle Attribute Recognition by Appearance: Computer Vision Methods for Vehicle Type, Make and Model Classification [0.9645196221785693]
  We survey a number of algorithms that identify vehicle properties ranging from coarse-grained level (vehicle type) to fine-grained level (vehicle make and model) We discuss two alternative approaches for these tasks, including straightforward classification and a more flexible metric learning method. We design a simulated real-world scenario for vehicle attribute recognition and present an experimental comparison of the two approaches.
  arXiv  Detail & Related papers  (2020-06-29T21:33:06Z)
• VehicleNet: Learning Robust Visual Representation for Vehicle Re-identification [116.1587709521173]
  We propose to build a large-scale vehicle dataset (called VehicleNet) by harnessing four public vehicle datasets. We design a simple yet effective two-stage progressive approach to learning more robust visual representation from VehicleNet. We achieve the state-of-art accuracy of 86.07% mAP on the private test set of AICity Challenge.
  arXiv  Detail & Related papers  (2020-04-14T05:06:38Z)
• The Devil is in the Details: Self-Supervised Attention for Vehicle Re-Identification [75.3310894042132]
  Self-supervised Attention for Vehicle Re-identification (SAVER) is a novel approach to effectively learn vehicle-specific discriminative features. We show that SAVER improves upon the state-of-the-art on challenging VeRi, VehicleID, Vehicle-1M and VERI-Wild datasets.
  arXiv  Detail & Related papers  (2020-04-14T02:24:47Z)
• Parsing-based View-aware Embedding Network for Vehicle Re-Identification [138.11983486734576]
  We propose a parsing-based view-aware embedding network (PVEN) to achieve the view-aware feature alignment and enhancement for vehicle ReID. The experiments conducted on three datasets show that our model outperforms state-of-the-art methods by a large margin.
  arXiv  Detail & Related papers  (2020-04-10T13:06:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.