Daniellesry/DKT-Mask-1-3B

Model Card for DKT

A transparent-obejct video segmentation model weight based on the learning strategy of Diffusion Knows Transparency: Repurposing Video Diffusion for Transparent Object Depth and Normal Estimation.

Usage

See the Github repository: DKT regarding installation instructions.

The model can then be used as follows:

from dkt.pipelines.pipelines import DKTPipeline
import os
from tools.common_utils import save_video

pipe = DKTPipeline()
demo_path = 'examples/1.mp4'
prediction = pipe(demo_path,prompt='transparent object segmentation')
save_dir = 'logs'
os.makedirs(save_dir, exist_ok=True)
output_path = os.path.join(save_dir, 'demo.mp4')
save_video(prediction['colored_depth_map'], output_path, fps=25)