Skip to content

Latest commit

 

History

History

EfficientSAM

README.md

Efficient Grounded-SAM

We're going to combine Grounding-DINO with efficient SAM variants for faster annotating.

Table of Contents

Installation

Efficient SAMs

Here's the list of Efficient SAM variants:

Title Intro Description Links
FastSAM The Fast Segment Anything Model(FastSAM) is a CNN Segment Anything Model trained by only 2% of the SA-1B dataset published by SAM authors. The FastSAM achieve a comparable performance with the SAM method at 50× higher run-time speed. [Github] [Demo]
MobileSAM MobileSAM performs on par with the original SAM (at least visually) and keeps exactly the same pipeline as the original SAM except for a change on the image encoder. Specifically, we replace the original heavyweight ViT-H encoder (632M) with a much smaller Tiny-ViT (5M). On a single GPU, MobileSAM runs around 12ms per image: 8ms on the image encoder and 4ms on the mask decoder. [Github]
Light-HQSAM Light HQ-SAM is based on the tiny vit image encoder provided by MobileSAM. We design a learnable High-Quality Output Token, which is injected into SAM's mask decoder and is responsible for predicting the high-quality mask. Instead of only applying it on mask-decoder features, we first fuse them with ViT features for improved mask details. Refer to Light HQ-SAM vs. MobileSAM for more details. [Github]
Efficient-SAM Segment Anything Model (SAM) has emerged as a powerful tool for numerous vision applications. However, the huge computation cost of SAM model has limited its applications to wider real-world applications. To address this limitation, we propose EfficientSAMs, light-weight SAM models that exhibit decent performance with largely reduced complexity. Our idea is based on leveraging masked image pretraining, SAMI, which learns to reconstruct features from SAM image encoder for effective visual representation learning. Further, we take SAMI-pretrained light-weight image encoders and mask decoder to build EfficientSAMs, and finetune the models on SA-1B for segment anything task. Refer to EfficientSAM arXiv for more details. [Github]

Run Grounded-FastSAM Demo

  • Firstly, download the pretrained Fast-SAM weight here

  • Run the demo with the following script:

cd Grounded-Segment-Anything

python EfficientSAM/grounded_fast_sam.py --model_path "./FastSAM-x.pt" --img_path "assets/demo4.jpg" --text "the black dog." --output "./output/"
  • And the results will be saved in ./output/ as:
Input Text Output
"The black dog."

Note: Due to the post process of FastSAM, only one box can be annotated at a time, if there're multiple box prompts, we simply save multiple annotate images to ./output now, which will be modified in the future release.

Run Grounded-MobileSAM Demo

  • Firstly, download the pretrained MobileSAM weight here

  • Run the demo with the following script:

cd Grounded-Segment-Anything

python EfficientSAM/grounded_mobile_sam.py --MOBILE_SAM_CHECKPOINT_PATH "./EfficientSAM/mobile_sam.pt" --SOURCE_IMAGE_PATH "./assets/demo2.jpg" --CAPTION "the running dog"
  • And the result will be saved as ./gronded_mobile_sam_anontated_image.jpg as:
Input Text Output
"The running dog"

Run Grounded-Light-HQSAM Demo

  • Firstly, download the pretrained Light-HQSAM weight here

  • Run the demo with the following script:

cd Grounded-Segment-Anything

python EfficientSAM/grounded_light_hqsam.py
  • And the result will be saved as ./gronded_light_hqsam_anontated_image.jpg as:
Input Text Output
"Bench"

Run Grounded-Efficient-SAM Demo

  • Download the pretrained EfficientSAM checkpoint from here and put it under Grounded-Segment-Anything/EfficientSAM

  • Run the demo with the following script:

cd Grounded-Segment-Anything

python EfficientSAM/grounded_efficient_sam.py
  • And the result will be saved as ./gronded_efficient_sam_anontated_image.jpg as:
Input Text Output
"Bench"