search

Tianfu18 / diff-feats-pose

[CVPR 2024 Highlight] PyTorch implementation of "Object Pose Estimation via the Aggregation of Diffusion Features"
29 stars 1 forks source link

Some questions about inference #2

Closed ran894694447 closed 2 months ago

ran894694447 commented 3 months ago

Thanks for your great work! I would like to ask if this model requires the object's mask during inference? What is the inference speed?Looking forward to your early reply!

Tianfu18 commented 2 months ago

Thank you for your attention to our work. We apologize for the delayed response. During inference, the object's mask is required, which is crucial for performance. As for inference speed, in our simple tests, on an A6000 GPU, the speed for the feature extraction part is about 10 frames per second.