Closed fatiherdogan01 closed 1 month ago
Doesn't selfie_multiclass_256x256 work on mobile as well? Personally I do not use deeplabv3, I preload MLKIT's model and once selfie_multiclass_256x256 is loaded I replace it. (See volcomix's implementation for mlkit's model: https://github.com/Volcomix/virtual-background)
I think it's equivalent to SelfieSegmenter (square)
You may have to use webgl2 instead of canvas2d for the bigger model to be efficient on mobile
Hi @fatiherdogan01,
We do not have a matrix comparison of the supported models. We suggest checking the overview page https://ai.google.dev/edge/mediapipe/solutions/vision/image_segmenter#models for model selection based on your image segmentation needs. You can also use a custom model in our API. Additionally, task benchmarks for all supported models on mobile device is available here https://ai.google.dev/edge/mediapipe/solutions/vision/image_segmenter#task-benchmarkse.
Thank you!!
This issue has been marked stale because it has no recent activity since 7 days. It will be closed if no further activity occurs. Thank you.
I'm now using this model https://github.com/xuebinqin/DIS and this is good.
Hi,
I'm using "selfie_multiclass_256x256" and "deeplab_v3" models for background removal. The "selfie multiclass 256x256" model works very well. But the "deeplabv3" model does not work well.
Does anyone have any suggestions for a lite model for mobile devices?