{"id":13935980,"url":"https://github.com/zhiqwang/sightseq","last_synced_at":"2025-08-14T03:31:50.992Z","repository":{"id":104220796,"uuid":"148242056","full_name":"zhiqwang/sightseq","owner":"zhiqwang","description":"Computer vision tools for fairseq, containing PyTorch implementation of text recognition and object detection","archived":false,"fork":false,"pushed_at":"2019-11-14T13:05:13.000Z","size":208,"stargazers_count":125,"open_issues_count":4,"forks_count":34,"subscribers_count":11,"default_branch":"master","last_synced_at":"2024-11-27T03:35:04.918Z","etag":null,"topics":["attention","crnn","ctc","densenet","faster-rcnn","image-captioning","mobilenet","object-detection","ocr","pytorch","scene-texts","text-recognition","transformer"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/zhiqwang.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2018-09-11T01:25:31.000Z","updated_at":"2024-11-21T09:05:02.000Z","dependencies_parsed_at":"2023-11-13T06:00:24.895Z","dependency_job_id":null,"html_url":"https://github.com/zhiqwang/sightseq","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhiqwang%2Fsightseq","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhiqwang%2Fsightseq/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhiqwang%2Fsightseq/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zhiqwang%2Fsightseq/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zhiqwang","download_url":"https://codeload.github.com/zhiqwang/sightseq/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":229795810,"owners_count":18125294,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["attention","crnn","ctc","densenet","faster-rcnn","image-captioning","mobilenet","object-detection","ocr","pytorch","scene-texts","text-recognition","transformer"],"created_at":"2024-08-07T23:02:15.849Z","updated_at":"2024-12-15T08:16:52.602Z","avatar_url":"https://github.com/zhiqwang.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# 🔭sightseq\n\nNow, Let's go **sight**seeing by vision and **seq**uence language multimodal around the deep learning world.\n\n### What's New:\n\n- **July 30, 2019:** Add faster rcnn models. And I rename this repo from *image-captioning* to *sightseq*, this is the last time I rename this repo, I promise.\n- **June 11, 2019:** I rewrite the text recognition part base on [fairseq](https://github.com/pytorch/fairseq). Stable version refer to branch [crnn](https://github.com/zhiqwang/image-captioning/tree/crnn), which provides pre-trained model checkpoints. Current branch is work in process. Very pleasure for suggestion and cooperation in the fairseq text recognition project.\n\n### Features:\nsightseq provides reference implementations of various deep learning tasks, including:\n\n- **Text Recognition**\n  - [Shi et al. (2015), CRNN: An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition](https://arxiv.org/abs/1507.05717)\n- **Object Detection**\n  - **_New_** [Ren et al. (2015), Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://arxiv.org/abs/1506.01497)\n\n**Additionally:**\n- All features of fairseq\n- Flexible to enable convolution layer, recurrent layer in CRNN\n- Positional Encoding of images\n\n# General Requirements and Installation\n- [PyTorch](http://pytorch.org/) (There is a [bug](https://github.com/pytorch/pytorch/pull/21244) in [nn.CTCLoss](https://pytorch.org/docs/master/nn.html#ctcloss) which is solved in nightly version)\n- Python version \u003e= 3.5\n- [Fairseq](https://github.com/pytorch/fairseq) version \u003e= 0.7.1\n- [torchvision](https://github.com/pytorch/vision) version \u003e= 0.3.0\n- For training new models, you'll also need an NVIDIA GPU and [NCCL](https://github.com/NVIDIA/nccl)\n\n# Pre-trained models and examples\n- [text recognition](examples/text_recognition)\n- [object detection](examples/object_detection)\n\n# License\nsightseq is MIT-licensed.\nThe license applies to the pre-trained models as well.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzhiqwang%2Fsightseq","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzhiqwang%2Fsightseq","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzhiqwang%2Fsightseq/lists"}