{"id":13627574,"url":"https://github.com/zunzhumu/S3FD","last_synced_at":"2025-04-17T00:31:58.841Z","repository":{"id":149174815,"uuid":"135055631","full_name":"zunzhumu/S3FD","owner":"zunzhumu","description":"S3FD_Mxnet","archived":false,"fork":false,"pushed_at":"2018-06-20T06:09:45.000Z","size":144,"stargazers_count":23,"open_issues_count":4,"forks_count":5,"subscribers_count":2,"default_branch":"master","last_synced_at":"2024-11-08T18:44:57.066Z","etag":null,"topics":["detector","face"],"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/zunzhumu.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}},"created_at":"2018-05-27T14:51:50.000Z","updated_at":"2023-09-08T17:41:01.000Z","dependencies_parsed_at":"2024-01-14T08:06:27.050Z","dependency_job_id":null,"html_url":"https://github.com/zunzhumu/S3FD","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/zunzhumu%2FS3FD","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zunzhumu%2FS3FD/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zunzhumu%2FS3FD/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zunzhumu%2FS3FD/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zunzhumu","download_url":"https://codeload.github.com/zunzhumu/S3FD/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":249292990,"owners_count":21245658,"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":["detector","face"],"created_at":"2024-08-01T22:00:35.884Z","updated_at":"2025-04-17T00:31:58.571Z","avatar_url":"https://github.com/zunzhumu.png","language":"Python","funding_links":[],"categories":["\u003ca name=\"Vision\"\u003e\u003c/a\u003e2. Vision"],"sub_categories":["2.5 Face Detection and Recognition"],"readme":"two stages strategies\\\nstage one: each gt matched anchor number:7\\\nstage one: each gt matched anchor number:2\\\nstage one: each gt matched anchor number:1\\\nstage one: each gt matched anchor number:1\\\nstage one: each gt matched anchor number:1\\\nstage one: each gt matched anchor number:3\\\nstage one: each gt matched anchor number:1\\\nstage one: each gt matched anchor number:9\\\nstage one: each gt matched anchor number:7\\\nstage one: each gt matched anchor number:9\\\nstage one: each gt matched anchor number:14\\\nthe ground truth number:11\\\nthe averge anchors matched number:5\\\ndeal with tiny and outer face\\\nstage two: each gt matched anchor number:7\\\nstage two: each gt matched anchor number:5\\\nstage two: each gt matched anchor number:5\\\nstage two: each gt matched anchor number:5\\\nstage two: each gt matched anchor number:5\\\nstage two: each gt matched anchor number:5\\\nstage two: each gt matched anchor number:5\\\nstage two: each gt matched anchor number:9\\\nstage two: each gt matched anchor number:7\\\nstage two: each gt matched anchor number:9\\\nstage two: each gt matched anchor number:14\n# S3FD: Single Shot Scale-invariant Face Detector\n\n### Getting started\n* You will need python modules: `cv2`, `matplotlib` and `numpy`.\nIf you use mxnet-python api, you probably have already got them.\nYou can install them via pip or package managers, such as `apt-get`:\n```\nsudo apt-get install python-opencv python-matplotlib python-numpy\n```\n## Note The scale compensation anchor matching strategy is written into multibox_target.cu\n* Copy multibox_target_operator/multibox_target.cc multibox_target.cu to mxnet/src/operator/contrib to cover original multibox_target.cc multibox_target.cu\n\n* Build MXNet: Follow the official instructions\n```\n\n### Train the model\nThis example only covers training on Wider Face dataset. Other datasets should\n* Download the converted pretrained `vgg16_reduced` model [here](https://github.com/zhreshold/mxnet-ssd/releases/download/v0.2-beta/vgg16_reduced.zip), unzip `.param` and `.json` files\ninto `model/` directory by default.\n* Download the Wider Face dataset, skip this step if you already have one.\n* Extra data in data/widerface/WIDER_train WIDER_val wider_face_split\n* Convert voc format: \ncd data/widerface\npython widerface_voc.py\n* Convert .rec data\npython tools/prepare_dataset.py --dataset widerface --set train --target ./data/train.lst\npython tools/prepare_dataset.py --dataset widerface --set val --target ./data/val.lst --shuffle False\n```\n* Start training:\n```\npython train.py\n\n### NOTE!!!!!!!!\n### By default,this example use data_shape=608, if you have enough GPU memory, you should set data_shape=640.\nI only have one GTX1080, so I don't have enough time to train.I use 0.001 learning rate for 7 epochs, the mAP achieved 64% in all validation set.You can try it at will.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzunzhumu%2FS3FD","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzunzhumu%2FS3FD","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzunzhumu%2FS3FD/lists"}