# FlipReID: Closing the Gap between Training and Inference in Person Re-Identification

arXiv: https://arxiv.org/abs/2105.05639 · HTML: https://mlsanity.com/paper/2105.05639/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 31 references that are in the map.
- First posted on arXiv 12 May 2021.

## Builds on (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Random Erasing Data Augmentation](https://mlsanity.com/paper/1708.04896/) (2017)
- [In Defense of the Triplet Loss for Person Re-Identification](https://mlsanity.com/paper/1703.07737/) (2017)
- [AutoAugment: Learning Augmentation Policies from Data](https://mlsanity.com/paper/1805.09501/) (2018)
- [Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking](https://mlsanity.com/paper/1609.01775/) (2016)
- [Deep Learning for Person Re-identification: A Survey and Outlook](https://mlsanity.com/paper/2001.04193/) (2020)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [Person Transfer GAN to Bridge Domain Gap for Person Re-Identification](https://mlsanity.com/paper/1711.08565/) (2017)
- [Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)](https://mlsanity.com/paper/1711.09349/) (2017)
- [ResNeSt: Split-Attention Networks](https://mlsanity.com/paper/2004.08955/) (2020)
- [Bag of Tricks and A Strong Baseline for Deep Person Re-identification](https://mlsanity.com/paper/1903.07071/) (2019)
- [Two at Once: Enhancing Learning and Generalization Capacities via IBN-Net](https://mlsanity.com/paper/1807.09441/) (2018)
- [Learning Discriminative Features with Multiple Granularities for Person Re-Identification](https://mlsanity.com/paper/1804.01438/) (2018)
- [Re-ranking Person Re-identification with k-reciprocal Encoding](https://mlsanity.com/paper/1701.08398/) (2017)
- [FastReID: A Pytorch Toolbox for General Instance Re-identification](https://mlsanity.com/paper/2006.02631/) (2020)
- [Improving Person Re-identification by Attribute and Identity Learning](https://mlsanity.com/paper/1703.07220/) (2017)
- [A Discriminatively Learned CNN Embedding for Person Re-identification](https://mlsanity.com/paper/1611.05666/) (2016)
- [A Siamese Long Short-Term Memory Architecture for Human Re-Identification](https://mlsanity.com/paper/1607.08381/) (2016)
- [Human-In-The-Loop Person Re-Identification](https://mlsanity.com/paper/1612.01345/) (2016)

## Built on by (most cited first, 16 of 16)

- [DINAR: Diffusion Inpainting of Neural Textures for One-Shot Human Avatars](https://mlsanity.com/paper/2303.09375/) (2023)
- [A Versatile Framework for Multi-scene Person Re-identification](https://mlsanity.com/paper/2403.11121/) (2024)
- [Offline-Online Associated Camera-Aware Proxies for Unsupervised Person Re-identification](https://mlsanity.com/paper/2201.05820/) (2022)
- [ReMix: Training Generalized Person Re-identification on a Mixture of Data](https://mlsanity.com/paper/2410.21938/) (2024)
- [Pose-dIVE: Pose-Diversified Augmentation with Diffusion Model for Person Re-Identification](https://mlsanity.com/paper/2406.16042/) (2024)
- [Specialized Re-Ranking: A Novel Retrieval-Verification Framework for Cloth Changing Person Re-Identification](https://mlsanity.com/paper/2210.03592/) (2022)
- [DynaMix: Generalizable Person Re-identification via Dynamic Relabeling and Mixed Data Sampling](https://mlsanity.com/paper/2511.19067/) (2025)
- [Motion-Aware Transformer For Occluded Person Re-identification](https://mlsanity.com/paper/2202.04243/) (2022)
- [3D Semantic MapNet: Building Maps for Multi-Object Re-Identification in 3D](https://mlsanity.com/paper/2403.13190/) (2024)
- [CORE-ReID: Comprehensive Optimization and Refinement through Ensemble fusion in Domain Adaptation for person re-identification](https://mlsanity.com/paper/2508.03064/) (2025)
- [Mind Your Clever Neighbours: Unsupervised Person Re-identification via Adaptive Clustering Relationship Modeling](https://mlsanity.com/paper/2112.01839/) (2021)
- [HBReID: Harder Batch for Re-identification](https://mlsanity.com/paper/2112.04761/) (2021)
- [Real-Time Online Unsupervised Domain Adaptation for Real-World Person Re-identification](https://mlsanity.com/paper/2306.03993/) (2023)
- [ISAR: A Benchmark for Single- and Few-Shot Object Instance Segmentation and Re-Identification](https://mlsanity.com/paper/2311.02734/) (2023)
- [A review of Recent Techniques for Person Re-Identification](https://mlsanity.com/paper/2509.22690/) (2025)
- [Reliability-Aware 3D Geometric Injection for Universal Person Re-identification](https://mlsanity.com/paper/2607.18863/) (2026)

## Abstract (arXiv)

Since neural networks are data-hungry, incorporating data augmentation in training is a widely adopted technique that enlarges datasets and improves generalization. On the other hand, aggregating predictions of multiple augmented samples (i.e., test-time augmentation) could boost performance even further. In the context of person re-identification models, it is common practice to extract embeddings for both the original images and their horizontally flipped variants. The final representation is the mean of the aforementioned feature vectors. However, such scheme results in a gap between training and inference, i.e., the mean feature vectors calculated in inference are not part of the training pipeline. In this study, we devise the FlipReID structure with the flipping loss to address this issue. More specifically, models using the FlipReID structure are trained on the original images and the flipped images simultaneously, and incorporating the flipping loss minimizes the mean squared error between feature vectors of corresponding image pairs. Extensive experiments show that our method brings consistent improvements. In particular, we set a new record for MSMT17 which is the largest person re-identification dataset. The source code is available at https://github.com/nixingyang/FlipReID.