TensorFlow.js 模型
探索预训练模型，将计算机视觉、自然语言处理 (NLP) 和其他常见机器学习任务添加到 Web 应用和基于浏览器的应用中。
身体
通过来自 MediaPipe 等项目的模型检测面部、手部和人体的关键点及姿势，这些模型已针对 JavaScript 和 Node.js 进行优化。
文本
借助 BERT 和其他 Transformer 编码器架构的功能在 Web 应用中实现 NLP。
[null,null,[],[],[],null,["# TensorFlow.js models\n====================\n\nExplore pre-trained models to add computer vision, natural language processing (NLP), and other common ML tasks to your web and browser-based applications. \n\nVision\n------\n\nAnalyze features in images and videos. Unlock new real-time experiences in the browser. \n[Image classification](https://github.com/tensorflow/tfjs-models/tree/master/mobilenet) \nClassify images with labels from the ImageNet database (MobileNet). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/mobilenet) \n[Object detection](https://github.com/tensorflow/tfjs-models/tree/master/coco-ssd) \nLocalize and identify multiple objects in a single image (Coco SSD). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/coco-ssd) \n[Semantic segmentation](https://github.com/tensorflow/tfjs-models/tree/master/deeplab) \nRun semantic segmentation in the browser (DeepLab). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/deeplab) \n\nBody\n----\n\nDetect key points and poses on the face, hands, and body with models from [MediaPipe](https://google.github.io/mediapipe/solutions/models) and beyond, optimized for JavaScript and Node.js. \n[Simple face detection](https://github.com/tensorflow/tfjs-models/tree/master/face-detection) \nDetect faces in images using a Single Shot Detector architecture with a custom encoder (Blazeface). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/face-detection) \n[Face landmark detection](https://github.com/tensorflow/tfjs-models/tree/master/face-landmarks-detection) \nPredict 486 3D facial landmarks to infer the approximate surface geometry of human faces. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/face-landmarks-detection) \n[Pose detection](https://github.com/tensorflow/tfjs-models/tree/master/pose-detection) \nUnified pose detection API for using one of three models that help detect atypical poses and fast body motions with real time performance. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/pose-detection) \n[Body segmentation](https://github.com/tensorflow/tfjs-models/tree/master/body-segmentation) \nSegment person(s) and body parts in real-time. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/body-segmentation) \n[Hand pose detection](https://github.com/tensorflow/tfjs-models/tree/master/hand-pose-detection) \nPalm detector and hand-skeleton finger tracking model. Predict 21 3D hand keypoints per detected hand. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/hand-pose-detection) \n[Portrait depth estimation](https://github.com/tensorflow/tfjs-models/tree/master/depth-estimation) \nEstimate a depth map for a single portrait image of a human. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/depth-estimation) \n\nText\n----\n\nEnable NLP in your web app using the power of BERT and other Transformer encoder architectures. \n[Natural language question answering](https://github.com/tensorflow/tfjs-models/tree/master/qna) \nAnswer questions based on the content of a given passage of text using BERT. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/qna) \n[Text toxicity detection](https://github.com/tensorflow/tfjs-models/tree/master/toxicity) \nScore the perceived impact a comment may have on a conversation, from \"Very toxic\" to \"Very healthy\" (Toxicity). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/toxicity) \n[Universal sentence encoder](https://github.com/tensorflow/tfjs-models/tree/master/universal-sentence-encoder) \nEncode text into embeddings for NLP tasks such as sentiment classification and textual similarity (Universal Sentence Encoder). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/universal-sentence-encoder) \n\nAudio\n-----\n\nClassify audio to detect sounds and trigger an action in your web app. \n[Speech command recognition](https://github.com/tensorflow/tfjs-models/tree/master/speech-commands) \nClassify 1-second audio snippets from the speech commands dataset (speech-commands). \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/speech-commands) \n\nGeneral\n-------\n\nFind more TensorFlow.js models that can be used out of the box. \n[KNN Classifier](https://github.com/tensorflow/tfjs-models/tree/master/knn-classifier) \nUtility to create a classifier using the K-Nearest-Neighbors algorithm. Can be used for transfer learning. \n[View code](https://github.com/tensorflow/tfjs-models/tree/master/knn-classifier) \n[Explore on GitHub](https://tfhub.dev/s?deployment-format=tfjs) \n\nGet started with TensorFlow.js\n------------------------------\n\n[Explore tutorials](/js/tutorials)"]]