mlsanityData to 7 Oct 2026

Bent & Broken Bicycles: Leveraging synthetic data for damaged object re-identification

arXiv 2304.07883 ↗First posted 16 Apr 2023cs.CV cs.GR cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Instance-level object re-identification is a fundamental computer vision task, with applications from image retrieval to intelligent monitoring and fraud detection. In this work, we propose the novel task of damaged object re-identification, which aims at distinguishing changes in visual appearance due to deformations or missing parts from subtle intra-class variations. To explore this task, we leverage the power of computer-generated imagery to create, in a semi-automatic fashion, high-quality synthetic images of the same bike before and after a damage occurs. The resulting dataset, Bent & Broken Bicycles (BBBicycles), contains 39,200 images and 2,800 unique bike instances spanning 20 different bike models. As a baseline for this task, we propose TransReI3D, a multi-task, transformer-based deep network unifying damage detection (framed as a multi-label classification task) with object re-identification. The BBBicycles dataset is available at https://huggingface.co/datasets/GrainsPolito/BBBicycles

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 42 of its 65 references that are in the map. First posted on arXiv 16 Apr 2023.

Builds on

42 of 65 references.

References · 42

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  3. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  4. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  5. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,401
  6. 2021Transformers in Vision: A Surveycited by 1,009
  7. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  8. 2016Playing for Data: Ground Truth from Computer Gamescited by 816
  9. 2017Person Transfer GAN to Bridge Domain Gap for Person Re-Identificationcited by 479
  10. 2021TransReID: Transformer-based Object Re-Identificationcited by 432
  11. 2016Learning from Simulated and Unsupervised Images through Adversarial Trainingcited by 287
  12. 2018Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objectscited by 237
30 more
  1. 2019Learning from Synthetic Data for Crowd Counting in the Wildcited by 148
  2. 2019Meta-Sim: Learning to Generate Synthetic Datasetscited by 128
  3. 2018Partial Adversarial Domain Adaptationcited by 127
  4. 2019Synthetic Data for Deep Learningcited by 113
  5. 2018Modeling Visual Context is Key to Augmenting Object Detection Datasetscited by 106
  6. 2018T2Net: Synthetic-to-Realistic Translation for Solving Single-Image Depth Estimation Taskscited by 70
  7. 2021MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?cited by 68
  8. 2016SIFT Meets CNN: A Decade Survey of Instance Retrievalcited by 66
  9. 2021Instance-level Image Retrieval using Reranking Transformerscited by 63
  10. 2021GiT: Graph Interactive Transformer for Vehicle Re-identificationcited by 49
  11. 2021AAformer: Auto-Aligned Transformer for Person Re-Identificationcited by 48
  12. 2021HAT: Hierarchical Aggregation Transformers for Person Re-identificationcited by 39
  13. 2019A survey of advances in vision-based vehicle re-identificationcited by 36
  14. 2020An Empirical Study of Multi-Task Learning on BERT for Biomedical Text Miningcited by 35
  15. 2021TransMatcher: Deep Image Matching Through Transformers for Generalizable Person Re-identificationcited by 32
  16. 2016Procedural Generation of Videos to Train Deep Action Recognition Networkscited by 31
  17. 2020PAMTRI: Pose-Aware Multi-Task Learning for Vehicle Re-Identification Using Highly Randomized Synthetic Datacited by 31
  18. 2020Deformation-Aware 3D Model Embedding and Retrievalcited by 30
  19. 2021BIKED: A Dataset for Computational Bicycle Design with Machine Learning Benchmarkscited by 27
  20. 2021Semantic-guided Pixel Sampling for Cloth-Changing Person Re-identificationcited by 22
  21. 2021Person Re-Identification with a Locally Aware Transformercited by 19
  22. 2021Hallucination In Object Detection -- A Study In Visual Part Verificationcited by 16
  23. 2021Discovering 3D Parts from Image Collectionscited by 13
  24. 2016UnrealStereo: Controlling Hazardous Factors to Analyze Stereo Visioncited by 8
  25. 2018An Anti-fraud System for Car Insurance Claim Based on Visual Evidencecited by 6
  26. 2019Benchmarking unsupervised near-duplicate image detectioncited by 5
  27. 2021Exceeding the Limits of Visual-Linguistic Multi-Task Learningcited by 3
  28. 2021Efficient large-scale image retrieval with deep feature orthogonality and Hybrid-Swin-Transformerscited by 3
  29. 2018DIY Human Action Data Set Generationcited by 2
  30. 2018Learning State Representations in Complex Systems with Multimodal Datacited by 1

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Addressing the Depth-of-Field Constraint: A New Paradigm for High Resolution Multi-Focus Image Fusioncited by 0