THU-MIG / RepViT

RepViT: Revisiting Mobile CNN From ViT Perspective [CVPR 2024] and RepViT-SAM: Towards Real-Time Segmenting Anything
https://arxiv.org/abs/2307.09283
Apache License 2.0
799 stars 60 forks source link

Question about the inference speed of RepViT-SAM on GPU device #30

Open mrbulb opened 11 months ago

mrbulb commented 11 months ago

Thank you for your great work! I notice that all the latency comparisons are conduct in Core ML platform. However it seems the MobileSAM report their latency on regular GPU platform according to their paper. I would like to ask if you have tested your model on conventional GPU devices, such as Nvidia's GPU platform, and how its performance compares to SAM and MobileSAM models.