TinySAM

TinySAM copied to clipboard

Hello, I'm trying to use TinySAM with the output of RTMDet as bounding box prompts. I came across the issue of the 'multimask_output' parameter in the 'predict' method of mask_predictor...

Riley-livingston

Loading Q-TinySAM

1

Hello! I came across two issues loading Q-TinySAM. First one is running `demo_quant.py` looking for normal checkpoints due to model type being `vit_t`, is this the proper type for loading...

merveenoyan

Tiny-Sam vs HQ-Sam

4

with four points, the tiny-sam does'nt segment correctly,while HQ-Sam works well even for pictures with great noises  

Meteplus

Hi, nice work it is. I'm tring your method to do some application and have some questions about the quantization. I have carefully looked at the code in demo_quan.py and...

ranpin

Which part is knowledge distillation

3

Hello, Author, Your paper is very nice, thanks for sharing. I have a confusion is that I can't find which part of the codes shows the knowledge distillation.

skill-diver

export the QTinySAM onnx file

1

 I want to export QTinySAM onnx file, but I met this question. Do you have a solution?

duxuan11

Finetune TinySAM on custom dataset

1

Hello, Im trying to fine-tune the mask decoder of tiny sam on a custom dataset while freezing the weights of the image_encoder and prompt_encoder. Im having an issue in my...

Riley-livingston

Questions about latency

2

 I have a question, does this time represent the runtime of a single image in SegEvery mode?

Greywan

Is the mask decoder weight inherited from the teacher models' decoder?

1

If so, in the full-stage knowledge distillation, the image encoder is randomly initialized, is the mask decoder finetuned at a smaller learning rate than the light weight image encoder? Is...

Vickeyhw

为什么量化模型的体积更大，推理速度更慢？

2

如题，使用设备为3060，对比相同输入的模型与量化模型，量化模型体积更大，推理速度更慢

RovingYoung