{"id":15009453,"url":"https://github.com/milkcat0904/temporal-segment-network-pytorch","last_synced_at":"2025-04-09T17:24:04.668Z","repository":{"id":141357027,"uuid":"115676123","full_name":"milkcat0904/temporal-segment-network-pytorch","owner":"milkcat0904","description":"tsn model for action recognition on pytorch","archived":false,"fork":false,"pushed_at":"2017-12-29T07:10:45.000Z","size":47344,"stargazers_count":17,"open_issues_count":3,"forks_count":4,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-03-23T19:23:10.807Z","etag":null,"topics":["action-recognition","python-3-6","pytorch","temporal-segment-networks"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/milkcat0904.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2017-12-29T02:04:35.000Z","updated_at":"2024-07-18T18:16:40.000Z","dependencies_parsed_at":null,"dependency_job_id":"64cd543d-974a-4b5b-9223-9a96869784ed","html_url":"https://github.com/milkcat0904/temporal-segment-network-pytorch","commit_stats":{"total_commits":9,"total_committers":1,"mean_commits":9.0,"dds":0.0,"last_synced_commit":"7abce4c9ae0a9ae35a20063d1a139d86ebf881b8"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/milkcat0904%2Ftemporal-segment-network-pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/milkcat0904%2Ftemporal-segment-network-pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/milkcat0904%2Ftemporal-segment-network-pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/milkcat0904%2Ftemporal-segment-network-pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/milkcat0904","download_url":"https://codeload.github.com/milkcat0904/temporal-segment-network-pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248075519,"owners_count":21043602,"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":["action-recognition","python-3-6","pytorch","temporal-segment-networks"],"created_at":"2024-09-24T19:25:24.047Z","updated_at":"2025-04-09T17:24:04.647Z","avatar_url":"https://github.com/milkcat0904.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Temporal Segment Networks\ntsn model for action recognition on pytorch\n\n1.编译环境 Build\n=\n\t\n\tbuild opencv 2.4.13 和 dense_flow 环境\n\n\t\t$ bash build_all.sh\n\t\t\n\t终端输出\"All tools built. Happy experimenting!\" 表示build成功\n\n2.下载数据集 Download dataset\n=\n实验数据集采用UCF-101，HMDB51\n* [ucf101][UCF101]\n* [hmdb51][HMDB51]\n\n3.提取光流 Extract optical flow\n=\n\t$ bash scripts/extract_optical_flow.sh DATASET_PATH OUT_PATH NUMBER_OF_WORKER\n\t\n\t\tDATASET_PATH : 数据集的地址\n\t\n\t\tOUT_PATH ：生成光流图的地址\n\t\n\t\tNUMBER_OF_WORKER ：工作的显卡数量，一般设置为2\n\n一个频频的光流图和frame和放在同一个文件夹下\n\t\n4.提取warped光流图 Etract warped flow\n=\n\n* 将`tools/build_of.py`的第89行`--flow_type`的默认值改为`warped_tv11`：\n\n```python\nparser.add_argument(\"--flow_type\", type=str, default='warped_tvl1', choices=['tvl1', 'warp_tvl1'])\n```\n\n* $ bash scripts/extract_optical_flow.sh DATASET_PATH OUT_PATH NUMBER_OF_WORKER\n\n5.生成标签 Label\n=\n* ucf101数据集标签：\n\n\t\t$ bash scripts/build_file_list.sh ucf101 FRAME_PATH\n\n* hmdb51数据集标签：\n\n\t\t$ bash scripts/build_file_list.sh hmdb51 FRAME_PATH\n\t\n\tFRAME_PATH：光流图（frame）的位置\n\t\n6.训练(Inception-BN) Training\n=\n* ucf101\n\t* rgb模型\n\t\n\t* flow模型\n\t\n\t* rgb-diff模型\n\t\n\t* warped flow模型\n\n* hmdb51\n\t* rgb模型\n\t\t\n\t\t\t$ python main.py hmdb51 RGB \u003chmdb51_rgb_train_list\u003e \u003chmdb51_rgb_val_list\u003e \n\t\t\n\t\t\t\t--arch BNInception --num_segments 3 \n\t\t\n\t\t\t\t--gd 20 --lr 0.001 --lr_steps 30 60 --epochs 80 \n\t\t\n\t\t\t\t-b 128 -j 8 --dropout 0.8 \n\t\t\n\t\t\t\t--snapshot_pref ucf101_bninception_ \n\t\t\n\t\t\t\t--gpus 0 1\n\t\t\n\t* flow模型\n\t\n\t* rgb-diff模型\n\t\n\t* warped flow模型\n\n[ucf101]:http://crcv.ucf.edu/data/UCF101.php\n[hmdb51]:http://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database/\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmilkcat0904%2Ftemporal-segment-network-pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmilkcat0904%2Ftemporal-segment-network-pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmilkcat0904%2Ftemporal-segment-network-pytorch/lists"}