Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval (CVPR 2023)
| Date | Stars |
|---|---|
| 2026-07-31 | 286 |
| 2026-08-06 | 288 |
Today
+2 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval [](https://github.com/anosorae/IRRA/blob/main/LICENSE) [](https://paperswithcode.com/sota/nlp-based-person-retrival-on-cuhk-pedes?p=cross-modal-implicit-relation-reasoning-and) Official PyTorch implementation of the paper Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval. (CVPR 2023) [arXiv](https://arxiv.org/abs/2303.12501) ## Updates - (3/23/2023) Add arXiv link for our paper. - (3/18/2023) Add download links of trained models and logs. - (3/17/2023) Ensure the reproducibility of our code. - (3/13/2023) Code released! ## Highlights The goal of this work is to enhance global text-to-image person retrieval performance, without requiring any additional supervision and inference cost. To achieve this, we utilize the full CLIP model as our feature extraction backbone. Additionally, we propose a novel cross-modal matching loss (SDM) and an Implicit Relation Reasoning module to mine fine-grained image-text relationships, enabling IRRA to learn more discriminative global image-text representations.  ## Usage ### Requirements we use single RTX3090 24G GPU for training and evaluation. ``` pytorch 1.9.0 torchvision 0.10.0 prettytable easydict ``` ### Prepare Datasets Download the CUHK-PEDES dataset from [here](https://github.com/ShuangLI59/Person-Search-with-Natural-Language-Description), ICFG-PEDES dataset from [here](https://github.com/zifyloo/SSAN) and RSTPReid dataset form [here](https://github.com/NjtechCVLab/RSTPReid-Dataset) Organize them in `your dataset root dir` folder as follows: ``` |-- your dataset root dir/ | |-- <CUHK-PEDES>/ | |-- imgs | |-- cam_a | |-- cam_b | |-- ... | |-- reid_raw.json | | |-- <ICFG-PEDES>/ | |-- imgs | |-- test | |-- train | |-- ICFG_PEDES.json | | |-- <RSTPReid>/ | |-- imgs | |-- data_captions.json ``` ## Training ```python python train.py \ --name iira \ --img_aug \ --batch_size 64 \ --MLM \ --loss_names 'sdm+mlm+id' \ --dataset_name 'CUHK-PEDES' \ --root_dir 'your dataset root dir' \ --num_epoch 60 ``` ## Testing ```python python test.py --config_file 'path/to/model_dir/configs.yaml' ``` ## IRRA on Text-to-Image Person Retrieval Results #### CUHK-PEDES dataset | Method | Backbone | Rank-1 | Rank-5 | Rank-10 | mAP | mINP | | :-------------: | :--------------: | :-------: | :-------: | :-------: | :-------: | :-------: | | CMPM/C | RN50/LSTM | 49.37 | - | 79.27 | - | - | | DSSL | RN50/BERT | 59.98 | 80.41 | 87.56 | - | - | | SSAN | RN50/LSTM | 61.37 | 80.15 | 86.73 | - | - | | Han et al. | RN101/Xformer | 64.08 | 81.73 | 88.19 | 60.08 | - | | LGUR | DeiT-Small/BERT | 65.25 | 83.12 | 89.00 | - | - | | IVT | ViT-B-16/BERT | 65.59 | 83.11 | 89.21 | - | - | | CFine | ViT-B-16/BERT | 69.57 | 85.93 | 91.15 | - | - | | **CLIP** | ViT-B-16/Xformer | 68.19 | 86.47 | 91.47 | 61.12 | 44.86 | | **IRRA (ours)** | ViT-B-16/Xformer | **73.38** | **89.93** | **93.71** | **66.13** | **50.24** | [Model & log for CUHK-PEDES](https://drive.google.com/file/d/1OBhFhpZpltRMZ88K6ceNUv4vZgevsFCW/view?usp=share_link) #### ICFG-PEDES dataset | Method | Rank-1 | Rank-5 | Rank-10 |
Excerpt of 6,349 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:af683d05194277f3, desc:text-to-image