SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2309.01420 · ICCV · 2023

Unified Pre-training with Pseudo Texts for Text-To-Image Person Re-identification

Jian Wang, Jingdong Wang, Xinyu Zhang, Zhiyin Shao, Changxing Ding

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 8 functions out of this paper's own repositories and ran 4 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
zhiyinshao-h/unipt canonical 4 of 7
ZhiyinShao-H/UniPT — 0 of 1
FunctionStatusWhere it lives
basic_clean Ran zhiyinshao-h/unipt/clip/simple_tokenizer.py
code served (permissive licence) · get_code("98f385d847636a3e")
get_pairs Ran zhiyinshao-h/unipt/clip/simple_tokenizer.py
code served (permissive licence) · get_code("d919ae32e5e4e616")
resize_pos_embed Ran zhiyinshao-h/unipt/clip/model.py
code served (permissive licence) · get_code("0570970b85b8d376")
whitespace_clean Ran zhiyinshao-h/unipt/clip/simple_tokenizer.py
code served (permissive licence) · get_code("9542161e9640b858")
ViLTransformerSS Not yet run ZhiyinShao-H/UniPT/vilt/modules/vilt_module.py
code served (permissive licence) · get_code("e7b6b004d47cf257")
build_model Not yet run zhiyinshao-h/unipt/clip/model.py
code served (permissive licence) · get_code("2acf7b3d245de310")
get_pretrained_tokenizer Not yet run zhiyinshao-h/unipt/vilt/datamodules/datamodule_base.py
code served (permissive licence) · get_code("6cc26ba2dedcb73e")
load Not yet run zhiyinshao-h/unipt/clip/clip.py
code served (permissive licence) · get_code("db64af1ba3479e8d")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The pre-training task is indispensable for the text-toimage person re-identification (T2I-ReID) task. However, there are two underlying inconsistencies between these two tasks that may impact the performance: i) Data inconsistency. A large domain gap exists between the generic images/texts used in public pre-trained models and the specific person data in the T2I-ReID task. This gap is especially severe for texts, as general textual data are usually unable to describe specific people in fine-grained detail. ii) Training inconsistency. The processes of pre-training of images and texts are independent, despite cross-modality learning being critical to T2I-ReID. To address the above issues, we present a new unified pre-training pipeline (UniPT) designed specifically for the T2I-ReID task. We first build a large-scale text-labeled person dataset "LUPerson-T", in which pseudo-textual descriptions of images are automatically generated by the CLIP paradigm using a divideconquer-combine strategy. Benefiting from this dataset, we then utilize a simple vision-and-language pre-training framework to explicitly align the feature space of the image and text modalities during pre-training. In this way, the pre-training task and the T2I-ReID task are made consistent with each other on both data and training levels. Without the need for any bells and whistles, our UniPT achieves competitive Rank-1 accuracy of, i.e., 68.50%, 60.09%, and 51.85% on CUHK-PEDES, ICFG-PEDES and RSTPReid, respectively. Both the LUPerson-T dataset and code are available at https://github.com/ZhiyinShao-H/UniPT.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2309.01420")
get_code_for_paper("2309.01420")
have("2309.01420")

Connect an agent — have() is free.