From a7225b938a5aff536dcdc4d48d7125964480f413 Mon Sep 17 00:00:00 2001 From: MediaPipe Team Date: Thu, 13 Aug 2020 15:02:55 -0400 Subject: [PATCH] Project import generated by Copybara. GitOrigin-RevId: aaca5c37abcf8b7a6c3c28804739afdbad46e704 --- README.md | 2 +- docs/index.md | 2 +- docs/solutions/pose.md | 21 +++++++++++---------- mediapipe/models/README.md | 12 ++++++------ 4 files changed, 19 insertions(+), 18 deletions(-) diff --git a/README.md b/README.md index e283484af..cfa0c2e77 100644 --- a/README.md +++ b/README.md @@ -88,7 +88,7 @@ run code search using ## Publications -* [BlazePose - On-device Real-time Body Pose Tracking](https://mediapipe.page.link/blazepose-blog) +* [BlazePose - On-device Real-time Body Pose Tracking](https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.html) in Google AI Blog * [MediaPipe Iris: Real-time Eye Tracking and Depth Estimation](https://ai.googleblog.com/2020/08/mediapipe-iris-real-time-iris-tracking.html) in Google AI Blog diff --git a/docs/index.md b/docs/index.md index 0c637e4e3..ca446d0f4 100644 --- a/docs/index.md +++ b/docs/index.md @@ -88,7 +88,7 @@ run code search using ## Publications -* [BlazePose - On-device Real-time Body Pose Tracking](https://mediapipe.page.link/blazepose-blog) +* [BlazePose - On-device Real-time Body Pose Tracking](https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.html) in Google AI Blog * [MediaPipe Iris: Real-time Eye Tracking and Depth Estimation](https://ai.googleblog.com/2020/08/mediapipe-iris-real-time-iris-tracking.html) in Google AI Blog diff --git a/docs/solutions/pose.md b/docs/solutions/pose.md index 92a8be981..7e8216662 100644 --- a/docs/solutions/pose.md +++ b/docs/solutions/pose.md @@ -5,7 +5,7 @@ parent: Solutions nav_order: 5 --- -# MediaPipe Pose +# MediaPipe BlazePose {: .no_toc } 1. TOC @@ -22,12 +22,13 @@ on top of the physical world in augmented reality. MediaPipe Pose is a ML solution for high-fidelity upper-body pose tracking, inferring 25 2D upper-body landmarks from RGB video frames utilizing our -[BlazePose](https://mediapipe.page.link/blazepose-blog) research. Current -state-of-the-art approaches rely primarily on powerful desktop environments for -inference, whereas our method achieves real-time performance on most modern -[mobile phones](#mobile), [desktops/laptops](#desktop), in [python](#python) and -even on the [web](#web). A variant of MediaPipe Pose that performs full-body -pose tracking on mobile phones will be included in an upcoming release of +[BlazePose](https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.html) +research. Current state-of-the-art approaches rely primarily on powerful desktop +environments for inference, whereas our method achieves real-time performance on +most modern [mobile phones](#mobile), [desktops/laptops](#desktop), in +[python](#python) and even on the [web](#web). A variant of MediaPipe Pose that +performs full-body pose tracking on mobile phones will be included in an +upcoming release of [ML Kit](https://developers.google.com/ml-kit/early-access/pose-detection). ![pose_tracking_upper_body_example.gif](../images/mobile/pose_tracking_upper_body_example.gif) | @@ -91,8 +92,8 @@ The landmark model currently included in MediaPipe Pose predicts the location of (x, y location and visibility), plus two virtual alignment keypoints. It shares the same architecture as the full-body version that predicts 33 landmarks, described in more detail in the -[BlazePose Google AI Blog](https://mediapipe.page.link/blazepose-blog) and in -this [paper](https://arxiv.org/abs/2006.10204). +[BlazePose Google AI Blog](https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.html) +and in this [paper](https://arxiv.org/abs/2006.10204). ![pose_tracking_upper_body_landmarks.png](../images/mobile/pose_tracking_upper_body_landmarks.png) | :------------------------------------------------------------------------------------------------: | @@ -168,7 +169,7 @@ Please refer to [these instructions](../index.md#mediapipe-on-the-web). ## Resources * Google AI Blog: - [BlazePose - On-device Real-time Body Pose Tracking](https://mediapipe.page.link/blazepose-blog) + [BlazePose - On-device Real-time Body Pose Tracking](https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.html) * Paper: [BlazePose: On-device Real-time Body Pose Tracking](https://arxiv.org/abs/2006.10204) ([presentation](https://youtu.be/YPpUOTRn5tA)) diff --git a/mediapipe/models/README.md b/mediapipe/models/README.md index 9222cd87d..9aa3c4367 100644 --- a/mediapipe/models/README.md +++ b/mediapipe/models/README.md @@ -5,7 +5,7 @@ * For back-facing cameras: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/face_detection_back.tflite) * [Model page](https://sites.google.com/corp/view/perception-cv4arvr/blazeface) * Paper: ["BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs"](https://arxiv.org/abs/1907.05047) - * [Model card](https://sites.google.com/corp/view/perception-cv4arvr/blazeface#h.p_21ojPZDx3cqq) + * [Model card](https://mediapipe.page.link/blazeface-mc) ### Face Mesh * Face detection: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/face_detection_front.tflite) (see above) @@ -14,12 +14,12 @@ * Paper: ["Real-time Facial Surface Geometry from Monocular Video on Mobile GPUs"](https://arxiv.org/abs/1907.06724) * [Google AI Blog post](https://ai.googleblog.com/2019/03/real-time-ar-self-expression-with.html) * [TensorFlow Blog post](https://blog.tensorflow.org/2020/03/face-and-hand-tracking-in-browser-with-mediapipe-and-tensorflowjs.html) - * [Model card](https://drive.google.com/file/d/1VFC_wIpw4O7xBOiTgUldl79d9LA-LsnA/view) + * [Model card](https://mediapipe.page.link/facemesh-mc) ### Hand Detection and Tracking * Palm detection: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/palm_detection.tflite), [TF.js model](https://tfhub.dev/mediapipe/handdetector/1) * 3D hand landmarks: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/hand_landmark.tflite), [TF.js model](https://tfhub.dev/mediapipe/handskeleton/1) - * [Google AI Blog post](https://mediapipe.page.link/handgoogleaiblog) + * [Google AI Blog post](https://ai.googleblog.com/2019/08/on-device-real-time-hand-tracking-with.html) * [TensorFlow Blog post](https://blog.tensorflow.org/2020/03/face-and-hand-tracking-in-browser-with-mediapipe-and-tensorflowjs.html) * [Model card](https://mediapipe.page.link/handmc) @@ -42,14 +42,14 @@ [BlazePose: On-device Real-time Body Pose Tracking](https://arxiv.org/abs/2006.10204) ([presentation](https://youtu.be/YPpUOTRn5tA)) * Google AI Blog: - [BlazePose - On-device Real-time Body Pose Tracking](https://mediapipe.page.link/blazepose-blog) + [BlazePose - On-device Real-time Body Pose Tracking](https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.html) * [Model card](https://mediapipe.page.link/blazepose-mc) ### Hair Segmentation * [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/hair_segmentation.tflite) * [Model page](https://sites.google.com/corp/view/perception-cv4arvr/hair-segmentation) * Paper: ["Real-time Hair segmentation and recoloring on Mobile GPUs"](https://arxiv.org/abs/1907.06740) - * [Model card](https://drive.google.com/file/d/1lPwJ8BD_-3UUor4LayQ0xpa_RIC_hoRh/view) + * [Model card](https://mediapipe.page.link/hairsegmentation-mc) ### Objectron (3D Object Detection) * Shoes: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/object_detection_3d_sneakers.tflite) @@ -64,6 +64,6 @@ * Up to 200 keypoints: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/knift_float.tflite) * Up to 400 keypoints: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/knift_float_400.tflite) * Up to 1000 keypoints: [TFLite model](https://github.com/google/mediapipe/tree/master/mediapipe/models/knift_float_1k.tflite) - * [Google Developers Blog post](https://mediapipe.page.link/knift) + * [Google Developers Blog post](https://developers.googleblog.com/2020/04/mediapipe-knift-template-based-feature-matching.html) * [Model card](https://mediapipe.page.link/knift-mc)