infinitered/nsfwjs

Run on every frame in Video

Open

#39 建立於 2019年3月4日

在 GitHub 查看
 (18 留言) (0 反應) (1 負責人)JavaScript (7,090 star) (507 fork)batch import
enhancementhacktoberfest-acceptedhelp wanted

描述

Much like the discussion over in #38 There's been a request to run on every frame in a video.

It's possible to do in JS like so: https://jsfiddle.net/bmartel/3h98gsvk/11/

Maybe a syntax like model.classifyVideo(video) or model.classifyAny(img|gif|video)

貢獻者指南

技術棧
javascripttensorflow
領域
machine learning
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
1-2 days
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
stale
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
JavaScript basicsfamiliarity with TensorFlow.jsunderstanding of video frame processing
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
50
研究方向
The issue requests adding video classification capability to the NSFWJS library, with a suggested API like `model.classifyVideo(video)`. There is a reference implementation in a JSFiddle (see issue body) that processes video frames using canvas and a 'draw' function. To implement, study how the existing `classify` method works for images, then extend it to accept a video element. The video should be sampled at a reasonable frame rate (e.g., requestAnimationFrame) and each frame passed to the model. Look at the discussion in the issue for additional context and any existing attempts. Key files to examine: the main model file (likely `index.js` or similar) and any image classification methods. Ensure the solution is efficient and handles cleanup of video resources.