MSMT17

Multi Scene Multi Time dataset for person re-id

Dataset Information
Modalities
Videos
Languages
Chinese
Introduced
2017
License
Unknown
Homepage
Official Website
Contents

Overview

MSMT17 is a multi-scene multi-time person re-identification dataset. The dataset consists of 180 hours of videos, captured by 12 outdoor cameras, 3 indoor cameras, and during 12 time slots. The videos cover a long period of time and present complex lighting variations, and it contains a large number of annotated identities, i.e., 4,101 identities and 126,441 bounding boxes.

The dataset can be requested at http://www.pkuvmc.com/dataset.html

Variants: MSMT17, Market-1501->MSMT17, DukeMTMC-reID->MSMT17, MSMT17->Market-1501

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

Task Model Paper Date
Person Re-Identification ReMix ReMix: Training Generalized Person Re-identification … 2024-10-29
Person Re-Identification BoT+UFFM+AMC Enhancing person re-identification via Uncertainty … 2024-05-02
Person Re-Identification CLIP-ReID Baseline + UFFM +AMC Enhancing person re-identification via Uncertainty … 2024-05-02
Person Re-Identification CA-Jaccard CA-Jaccard: Camera-aware Jaccard Distance for … 2023-11-17
Person Re-Identification PCL-CLIP (L_pcl) Prototypical Contrastive Learning-based CLIP Fine-tuning … 2023-10-26
Person Re-Identification PCL-CLIP (L_pcl+L_id) Prototypical Contrastive Learning-based CLIP Fine-tuning … 2023-10-26
Person Re-Identification ProNet++ (ResNet50+RK) Rethinking Person Re-identification from a … 2023-08-21
Person Re-Identification ProNet++ Rethinking Person Re-identification from a … 2023-08-21
Person Re-Identification SOLIDER (without re-ranking) Beyond Appearance: a Semantic Controllable … 2023-03-30
Person Re-Identification SOLIDER (with re-ranking) Beyond Appearance: a Semantic Controllable … 2023-03-30
Person Re-Identification MSINet (2.3M w/o RK) MSINet: Twins Contrastive Search of … 2023-03-13
Person Re-Identification UniHCP (finetune) UniHCP: A Unified Model for … 2023-03-06
Person Re-Identification DiP (without RK) DiP: Learning Discriminative Implicit Parts … 2022-12-24
Person Re-Identification CLIP-ReID (without re-ranking) CLIP-ReID: Exploiting Vision-Language Model for … 2022-11-25
Person Re-Identification CLIP-ReID (with re-ranking) CLIP-ReID: Exploiting Vision-Language Model for … 2022-11-25
Person Re-Identification Weakly Supervised Pre-training (ResNet50+MGN) Large-Scale Pre-training for Person Re-identification … 2022-03-30
Person Re-Identification TransReID-SSL (ViT-B without RK) Self-Supervised Pre-Training for Transformer-Based Person … 2021-11-23
Person Re-Identification TransReID-SSL (without RK) Self-Supervised Pre-Training for Transformer-Based Person … 2021-11-23
Person Re-Identification LDS (ResNet50+RK) Learning to Disentangle Scenes for … 2021-11-10
Person Re-Identification CAL(ResNet50) Counterfactual Attention Learning for Fine-Grained … 2021-08-19

Research Papers

Recent papers with results on this dataset:

External Links: