search

open-mmlab / playground

A central hub for gathering and showcasing amazing projects that extend OpenMMLab with SAM and other exciting features.
Apache License 2.0
1.1k stars 122 forks source link

如何使用label_anything只做分割标注？根据教程标注出来的是目标检测加分割？ #153

Open wangqiuji1 opened 1 year ago

wangqiuji1 commented 1 year ago
1693062495884
dky233 commented 1 year ago

只输出多边形标注就可以了，如下： label-studio-ml start sam --port 8003 --with \ model_name=mobile_sam \ sam_config=vit_t \ sam_checkpoint_file=./mobile_sam.pt \ device=cpu out_poly=True

JimmyMa99 commented 1 year ago

教程做的是实例分割（mask-rcnn），我们推荐使用rle格式输出的coco，也就是脚本输出的rle coco格式

JimmyMa99 commented 11 months ago

问题已经解决，请看 https://github.com/open-mmlab/playground/pull/164

获取Outlook for Androidhttps://aka.ms/AAb9ysg

From: Hebei Li @.> Sent: Sunday, October 15, 2023 2:58:33 PM To: open-mmlab/playground @.> Cc: Ma Zhiming @.>; Comment @.> Subject: Re: [open-mmlab/playground] 如何使用label_anything只做分割标注？根据教程标注出来的是目标检测加分割？ (Issue #153)

我发现hq-sam好像支持不了， Unexpected key(s) in state_dict: "mask_decoder.hf_token.weight", "mask_decoder.hf_mlp.layers.0.weight", "mask_decoder.hf_mlp.layers.0.bias", "mask_decoder.hf_mlp.layers.1.weight", "mask_decoder.hf_mlp.layers.1.bias", "mask_decoder.hf_mlp.layers.2.weight", "mask_decoder.hf_mlp.layers.2.bias", "mask_decoder.compress_vit_feat.0.weight", "mask_decoder.compress_vit_feat.0.bias", "mask_decoder.compress_vit_feat.1.weight", "mask_decoder.compress_vit_feat.1.bias", "mask_decoder.compress_vit_feat.3.weight", "mask_decoder.compress_vit_feat.3.bias", "mask_decoder.embedding_encoder.0.weight", "mask_decoder.embedding_encoder.0.bias", "mask_decoder.embedding_encoder.1.weight", "mask_decoder.embedding_encoder.1.bias", "mask_decoder.embedding_encoder.3.weight", "mask_decoder.embedding_encoder.3.bias", "mask_decoder.embedding_maskfeature.0.weight", "mask_decoder.embedding_maskfeature.0.bias", "mask_decoder.embedding_maskfeature.1.weight", "mask_decoder.embedding_maskfeature.1.bias", "mask_decoder.embedding_maskfeature.3.weight", "mask_decoder.embedding_maskfeature.3.bias".

― Reply to this email directly, view it on GitHubhttps://github.com/open-mmlab/playground/issues/153#issuecomment-1763298031, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AYGOLCDP6QJ5XHE47WJ7WI3X7OCRTANCNFSM6AAAAAA37U6OT4. You are receiving this because you commented.Message ID: @.***>