Search results

Face Landmarker has shorter detection range on iOS compared to Web

Status: Open.

#5866 In google-ai-edge/mediapipe;

· rcanbaba opened on Feb 21, 2025

Running image segmenter on codepen gives different results than running image segmenter on mediapipe studio on ios chrome

platform:iosMediaPipe IOS issuesMediaPipe IOS issues

task:image segmentationIssues related to image segmentation: Locate objects and create image masks with labelsIssues related to image segmentation: Locate objects and create image masks with labels

type:performanceExecution Time and memory heap, stackoverflow and garbage collection relatedExecution Time and memory heap, stackoverflow and garbage collection related

type:supportGeneral questionsGeneral questions

Status: Open.

#5862 In google-ai-edge/mediapipe;

· kaneishere opened on Feb 18, 2025

Error occurs while minifying with R8 (google auto value dependencies conflict with some other packages: Firebase,Camera2, etc.)

platform::androidAndroid SolutionsAndroid Solutions

stat:awaiting responseWaiting for user responseWaiting for user response

task:face detectionIssues related to Face Detection: Identify faces in images and videoIssues related to Face Detection: Identify faces in images and video

type:bugBug in the Source Code of MediaPipe SolutionBug in the Source Code of MediaPipe Solution

Status: Open.

#5859 In google-ai-edge/mediapipe;

· JenyaKirmiza opened on Feb 16, 2025

No module named 'mediapipe.python._framework_bindings'

os:linux-non-armIssues on linux distributions which run on x86-64 architecture. DOES NOT include ARM devices.Issues on linux distributions which run on x86-64 architecture. DOES NOT include ARM devices.

os:macOSIssues on MacOSIssues on MacOS

platform:pythonMediaPipe Python issuesMediaPipe Python issues

type:modelmakerIssues related to creation of custom on-device ML solutionsIssues related to creation of custom on-device ML solutions

Status: Open.

#5858 In google-ai-edge/mediapipe;

· PranayLohia opened on Feb 13, 2025

Add Tongue Tracking to MediaPipe Face Mesh and Blendshpaes

stat:awaiting googlerWaiting for Google Engineer's ResponseWaiting for Google Engineer's Response

task:face landmarkerIssues related to Face Landmarker: Identify facial features for visual effects and avatars.Issues related to Face Landmarker: Identify facial features for visual effects and avatars.

type:featureEnhancement in the New Functionality or Request for a New SolutionEnhancement in the New Functionality or Request for a New Solution

Status: Open.

#5857 In google-ai-edge/mediapipe;

· gb2111 opened on Feb 12, 2025

ModuleNotFoundError: No module named 'keras.src.engine'

os:macOSIssues on MacOSIssues on MacOS

stale

stat:awaiting responseWaiting for user responseWaiting for user response

task:object detectionIssues related to Object detection: Track and label objects in images and video.Issues related to Object detection: Track and label objects in images and video.

type:bugBug in the Source Code of MediaPipe SolutionBug in the Source Code of MediaPipe Solution

type:modelmakerIssues related to creation of custom on-device ML solutionsIssues related to creation of custom on-device ML solutions

Status: Open.

#5856 In google-ai-edge/mediapipe;

· plowman opened on Feb 12, 2025

GPU or TPU not detected with GPU/TPU runtime post mediapipe & mediapipe_model_maker installation

gpuMediaPipe GPU related issuesMediaPipe GPU related issues

os:linux-non-armIssues on linux distributions which run on x86-64 architecture. DOES NOT include ARM devices.Issues on linux distributions which run on x86-64 architecture. DOES NOT include ARM devices.

os:macOSIssues on MacOSIssues on MacOS

type:bugBug in the Source Code of MediaPipe SolutionBug in the Source Code of MediaPipe Solution

type:modelmakerIssues related to creation of custom on-device ML solutionsIssues related to creation of custom on-device ML solutions

Status: Open.

#5855 In google-ai-edge/mediapipe;

· PranayLohia opened on Feb 11, 2025

How to get NormalizedLandmarkList of facemesh with C++ interface?

legacy:face meshIssues related to Face MeshIssues related to Face Mesh

os:linux-non-armIssues on linux distributions which run on x86-64 architecture. DOES NOT include ARM devices.Issues on linux distributions which run on x86-64 architecture. DOES NOT include ARM devices.

platform:c++Issues specific to C++ framework in mediapipeIssues specific to C++ framework in mediapipe

type:supportGeneral questionsGeneral questions

Status: Open.

#5853 In google-ai-edge/mediapipe;

· yonger001 opened on Feb 10, 2025

No module named 'mediapipe.tasks.vision'

os:windowsMediaPipe issues on WindowsMediaPipe issues on Windows

platform:pythonMediaPipe Python issuesMediaPipe Python issues

stale

stat:awaiting responseWaiting for user responseWaiting for user response

task::allAll tasks of MediaPipeAll tasks of MediaPipe

type:bugBug in the Source Code of MediaPipe SolutionBug in the Source Code of MediaPipe Solution

Status: Open.

#5852 In google-ai-edge/mediapipe;

· DarshanJain-07 opened on Feb 9, 2025

Question about how FaceLandmarker works

legacy:face meshIssues related to Face MeshIssues related to Face Mesh

type:supportGeneral questionsGeneral questions

Status: Open.

#5851 In google-ai-edge/mediapipe;

· Moki26 opened on Feb 9, 2025

Customized tflike model can't work in iphone app

platform:iosMediaPipe IOS issuesMediaPipe IOS issues

task:hand landmarkerIssues related to hand landmarker: Identify and track hands and fingersIssues related to hand landmarker: Identify and track hands and fingers

type:modelmakerIssues related to creation of custom on-device ML solutionsIssues related to creation of custom on-device ML solutions

Status: Open.

#5850 In google-ai-edge/mediapipe;

· gzhhong opened on Feb 8, 2025

Add possibility to update FaceLandmarker Task parameter

platform:pythonMediaPipe Python issuesMediaPipe Python issues

stat:awaiting googlerWaiting for Google Engineer's ResponseWaiting for Google Engineer's Response

task:face landmarkerIssues related to Face Landmarker: Identify facial features for visual effects and avatars.Issues related to Face Landmarker: Identify facial features for visual effects and avatars.

type:featureEnhancement in the New Functionality or Request for a New SolutionEnhancement in the New Functionality or Request for a New Solution

Status: Open.

#5847 In google-ai-edge/mediapipe;

· cansik opened on Feb 3, 2025