search

hkchengrex / XMem

[ECCV 2022] XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model
https://hkchengrex.com/XMem/
MIT License
1.72k stars 191 forks source link

Your model trained with 4 GPU(12gb) machine does not perform well. #108

Closed longmalongma closed 1 year ago

longmalongma commented 1 year ago

Your model trained with 4 GPU(12gb) machine does not perform well. Can you provide a set of parameters suitable for 4-gpu (12gb)?

hkchengrex commented 1 year ago

I am not sure what causes the performance variation (https://github.com/hkchengrex/XMem/issues/71) so I cannot even start tuning it for the 4-GPU setting.