{"id":19146459,"url":"https://github.com/kahsolt/emotionroi-ex","last_synced_at":"2026-06-12T00:30:29.218Z","repository":{"id":214291337,"uuid":"736122910","full_name":"Kahsolt/EmotionROI-ex","owner":"Kahsolt","description":"Unofficial implementation of the EmotionROI essay and extended applications","archived":false,"fork":false,"pushed_at":"2024-04-20T06:41:37.000Z","size":2167,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-01-03T16:28:16.038Z","etag":null,"topics":["emotion-analysis","emotion-classification","region-of-interest"],"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/Kahsolt.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":"2023-12-27T03:44:26.000Z","updated_at":"2024-04-20T06:41:41.000Z","dependencies_parsed_at":null,"dependency_job_id":"2a562825-a528-4436-8a7d-f7cea55912cf","html_url":"https://github.com/Kahsolt/EmotionROI-ex","commit_stats":null,"previous_names":["kahsolt/emotionroi-ex"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Kahsolt%2FEmotionROI-ex","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Kahsolt%2FEmotionROI-ex/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Kahsolt%2FEmotionROI-ex/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Kahsolt%2FEmotionROI-ex/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Kahsolt","download_url":"https://codeload.github.com/Kahsolt/EmotionROI-ex/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240229935,"owners_count":19768588,"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":["emotion-analysis","emotion-classification","region-of-interest"],"created_at":"2024-11-09T07:44:12.223Z","updated_at":"2026-06-12T00:30:28.948Z","avatar_url":"https://github.com/Kahsolt.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# EmotionROI-ex\n\n    Unofficial implementation of the EmotionROI essay and extended applications\n\n----\n\nℹ The pretrained weights can be found [here](https://pan.quark.cn/s/13e59e043b1c) :)\n\nReproduction the **FCNEL** model in essay [WHERE DO EMOTIONS COME FROM? PREDICTING THE EMOTION STIMULI MAP](http://chenlab.ece.cornell.edu/Publication/Kuan-Chuan/ICIP16_EmotionROI.pdf), and extended applications.\n\nThe results reported in the EmotionROI/FCNEL essay:\n\n![FCNEL](img/FCNEL.png)\n\n\n### Reproduce FCNEL for ROI prediction\n\n⚠ Can NOT reproduce essay results exactly, here're the implementation differences:\n\n- The essay uses `FCN-32s + AlexNet`, we use `FCN-8s + ResNet50`\n- The essay uses `lr=1e-8` to refine `20` epochs, which is two small and doesn't have any effect in our setting, we instead use `lr=1e-2` to train `100` epochs with `CosineAnnealingLR`\n- The essay might use the default `Bilinear` for output upsacling, we instead use `Nearest-Exact` to simulate `FCN-32s` resolution\n\n![FCNEL-infer](img/FCNEL-infer.png)  \n![FCNEL-loss](img/FCNEL-loss.png)\n\n\n### Modifyied EmoFCNEL for ROI prediction with emotion condition\n\nℹ Predict ROI with any emotion label (except for `sadness`) results into nearly the same output...  \nℹ The ROI annotations are probably NOT related to certain emotion, but only indicates **human attention focus**  \n\n![EmoFCNEL-infer](img/EmoFCNEL-infer.png)\n![EmoFCNEL-loss](img/EmoFCNEL-loss.png)\n![EmoFCNEL-embed](img/EmoFCNEL-embed.png)\n\n\n### Predict emotion directly from ROI annotation\n\n⚠ Mission impossible :(\n\n| Model | Train acc. | Test acc. |\n| :-: | :-: | :-: |\n| Resnet50 | 24.098% | 17.172% |\n\n![RIO_clf](img/RIO_clf-acc.png)\n\n\n### references\n\n- essay\n  - info page: [http://chenlab.ece.cornell.edu/publications.html](http://chenlab.ece.cornell.edu/publications.html)\n  - download link: [http://chenlab.ece.cornell.edu/Publication/Kuan-Chuan/ICIP16_EmotionROI.pdf](http://chenlab.ece.cornell.edu/Publication/Kuan-Chuan/ICIP16_EmotionROI.pdf)\n- EmotionROI dataset\n  - info page: [http://chenlab.ece.cornell.edu/downloads.html](http://chenlab.ece.cornell.edu/downloads.html)\n  - download link: [http://chenlab.ece.cornell.edu/people/kuanchuan/publications/EmotionROI.zip](http://chenlab.ece.cornell.edu/people/kuanchuan/publications/EmotionROI.zip)\n\n----\nby Armit\n2023/12/27 \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkahsolt%2Femotionroi-ex","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkahsolt%2Femotionroi-ex","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkahsolt%2Femotionroi-ex/lists"}