{"id":13528149,"url":"https://github.com/Karel911/TRACER","last_synced_at":"2025-04-01T11:31:00.819Z","repository":{"id":39663126,"uuid":"438542837","full_name":"Karel911/TRACER","owner":"Karel911","description":"TRACER: Extreme Attention Guided Salient Object Tracing Network (AAAI 2022) implementation in PyTorch","archived":false,"fork":false,"pushed_at":"2024-09-11T00:18:49.000Z","size":10097,"stargazers_count":195,"open_issues_count":14,"forks_count":41,"subscribers_count":5,"default_branch":"main","last_synced_at":"2024-11-02T13:34:19.554Z","etag":null,"topics":["aaai-2022","aaai2022","attention","attention-mechanism","background-removal","image-segmentation","pytorch","pytorch-implementation","salient-object-detection"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Karel911.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":"2021-12-15T07:56:23.000Z","updated_at":"2024-09-11T00:18:52.000Z","dependencies_parsed_at":"2024-01-13T22:23:00.579Z","dependency_job_id":"d39159bf-9203-4615-85dd-3db572987091","html_url":"https://github.com/Karel911/TRACER","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Karel911%2FTRACER","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Karel911%2FTRACER/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Karel911%2FTRACER/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Karel911%2FTRACER/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Karel911","download_url":"https://codeload.github.com/Karel911/TRACER/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246631698,"owners_count":20808738,"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":["aaai-2022","aaai2022","attention","attention-mechanism","background-removal","image-segmentation","pytorch","pytorch-implementation","salient-object-detection"],"created_at":"2024-08-01T06:02:14.893Z","updated_at":"2025-04-01T11:31:00.364Z","avatar_url":"https://github.com/Karel911.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# TRACER: Extreme Attention Guided Salient Object Tracing Network\n\nThis paper was accepted at AAAI 2022 SA poster session. [[pdf]](https://arxiv.org/abs/2112.07380)    \n\n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/tracer-extreme-attention-guided-salient/salient-object-detection-on-duts-te)](https://paperswithcode.com/sota/salient-object-detection-on-duts-te?p=tracer-extreme-attention-guided-salient)  \n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/tracer-extreme-attention-guided-salient/salient-object-detection-on-dut-omron)](https://paperswithcode.com/sota/salient-object-detection-on-dut-omron?p=tracer-extreme-attention-guided-salient)  \n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/tracer-extreme-attention-guided-salient/salient-object-detection-on-hku-is)](https://paperswithcode.com/sota/salient-object-detection-on-hku-is?p=tracer-extreme-attention-guided-salient)  \n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/tracer-extreme-attention-guided-salient/salient-object-detection-on-ecssd)](https://paperswithcode.com/sota/salient-object-detection-on-ecssd?p=tracer-extreme-attention-guided-salient)  \n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/tracer-extreme-attention-guided-salient/salient-object-detection-on-pascal-s)](https://paperswithcode.com/sota/salient-object-detection-on-pascal-s?p=tracer-extreme-attention-guided-salient) \n\n![alt text](img/Poster.png)\n\n## Updates\n[09/06/2022] Demo has been released on [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1ZGbxozNHsvnOiywYZARGXr_CvvE6jIFh?usp=sharing/) [Try it now!](https://colab.research.google.com/drive/1ZGbxozNHsvnOiywYZARGXr_CvvE6jIFh?usp=sharing/)  \n\n\n\n\n[06/17/2022] Now, fast inference mode offers a salient object result with the mask.  \nWe have improved a result quality of salient object as follows.  \nYou can get the more clear salient object by tuning the [threshold](https://github.com/Karel911/TRACER/blob/main/inference.py/#L71).\n![img](https://user-images.githubusercontent.com/46666862/174314293-da0f9c73-6735-4012-9655-e340dfc251c1.png)\nWe will release initializing TRACER with a version of pre-trained TE-x.\n\n[04/20/2022] We update a pipeline for custom dataset inference w/o measuring.\n* Run **main.py** scripts.\n\u003cpre\u003e\u003ccode\u003e\nTRACER\n├── data\n│   ├── custom_dataset\n│   │   ├── sample_image1.png\n│   │   ├── sample_image2.png\n      .\n      .\n      .\n\n# For testing TRACER with pre-trained model (e.g.)  \npython main.py inference --dataset custom_dataset/ --arch 7 --img_size 640 --save_map True\n\u003c/code\u003e\u003c/pre\u003e\n\n\n## Datasets\nAll datasets are available in public.\n* Download the DUTS-TR and DUTS-TE from [Here](http://saliencydetection.net/duts/#org3aad434)\n* Download the DUT-OMRON from [Here](http://saliencydetection.net/dut-omron/#org96c3bab)\n* Download the HKU-IS from [Here](https://sites.google.com/site/ligb86/hkuis)\n* Download the ECSSD from [Here](https://www.cse.cuhk.edu.hk/leojia/projects/hsaliency/dataset.html)\n* Download the PASCAL-S from [Here](http://cbs.ic.gatech.edu/salobj/)\n* Download the edge GT from [Here](https://drive.google.com/file/d/1FX2RVeMxPgmSALQUSKhdiNrzf_HxA1o9/view?usp=sharing).\n\n## Data structure\n\u003cpre\u003e\u003ccode\u003e\nTRACER\n├── data\n│   ├── DUTS\n│   │   ├── Train\n│   │   │   ├── images\n│   │   │   ├── masks\n│   │   │   ├── edges\n│   │   ├── Test\n│   │   │   ├── images\n│   │   │   ├── masks\n│   ├── DUT-O\n│   │   ├── Test\n│   │   │   ├── images\n│   │   │   ├── masks\n│   ├── HKU-IS\n│   │   ├── Test\n│   │   │   ├── images\n│   │   │   ├── masks\n      .\n      .\n      .\n\u003c/code\u003e\u003c/pre\u003e\n\n## Requirements\n* Python \u003e= 3.7.x\n* Pytorch \u003e= 1.8.0\n* albumentations \u003e= 0.5.1\n* tqdm \u003e=4.54.0\n* scikit-learn \u003e= 0.23.2\n\n## Run\n* Run **main.py** scripts.\n\u003cpre\u003e\u003ccode\u003e\n# For training TRACER-TE0 (e.g.)\npython main.py train --arch 0 --img_size 320\n\n# For testing TRACER with pre-trained model (e.g.)  \npython main.py test --exp_num 0 --arch 0 --img_size 320\n\u003c/code\u003e\u003c/pre\u003e\n* Pre-trained models of TRACER are available at [here](https://github.com/Karel911/TRACER/releases/tag/v1.0)\n* Change the model name as 'best_model.pth' and put the weights to the path 'results/DUTS/TEx_0/best_model.pth'  \n  (here, the x means the model scale e.g., 0 to 7).\n* Input image sizes for each model are listed belows.\n\n## Configurations\n--arch: EfficientNet backbone scale: TE0 to TE7.  \n--frequency_radius: High-pass filter radius in the MEAM.  \n--gamma: channel confidence ratio \\gamma in the UAM.   \n--denoise: Denoising ratio d in the OAM.  \n--RFB_aggregated_channel: # of channels in receptive field blocks.  \n--multi_gpu: Multi-GPU learning options.  \n--img_size: Input image resolution.  \n--save_map: Options saving predicted mask.  \n\n\u003ctable\u003e\n\u003cthead\u003e\n  \u003ctr\u003e\n    \u003cth\u003eModel\u003c/th\u003e\n    \u003cth\u003eImg size\u003c/th\u003e\n  \u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-0 ~ 1\u003c/td\u003e\n        \u003ctd\u003e320\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-2\u003c/td\u003e\n        \u003ctd\u003e352\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-3\u003c/td\u003e\n        \u003ctd\u003e384\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-4\u003c/td\u003e\n        \u003ctd\u003e448\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-5\u003c/td\u003e\n        \u003ctd\u003e512\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-6\u003c/td\u003e\n        \u003ctd\u003e576\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTRACER-Efficient-7\u003c/td\u003e\n        \u003ctd\u003e640\u003c/td\u003e\n    \u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\n## Citation\n\u003cpre\u003e\u003ccode\u003e\n@article{lee2021tracer,\n  title={TRACER: Extreme Attention Guided Salient Object Tracing Network},\n  author={Lee, Min Seok and Shin, WooSeok and Han, Sung Won},\n  journal={arXiv preprint arXiv:2112.07380},\n  year={2021}\n}\n\u003c/code\u003e\u003c/pre\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FKarel911%2FTRACER","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FKarel911%2FTRACER","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FKarel911%2FTRACER/lists"}