how to use 27 frames

[parkjin0903]

Thank you for providing the inference code. I am very interested in your paper. According to the paper and code, it seems like using 351 frames apparently. I tried to modify vis.py code to use 27 frames. Unfortunately I faced a hardship. The following is the error code that I got. Could you give me a instruction?

---

size mismatch for Transformer.pos_embedding: copying a param with shape torch.Size([1, 351, 256]) from checkpoint, the shape in current model is torch.Size([1, 27, 256]). size mismatch for Transformer_reduce.model.pos_embedding_1: copying a param with shape torch.Size([1, 351, 256]) from checkpoint, the shape in current model is torch.Size([1, 27, 256]). size mismatch for Transformer_reduce.model.pos_embedding_2: copying a param with shape torch.Size([1, 351, 256]) from checkpoint, the shape in current model is torch.Size([1, 27, 256]). size mismatch for Transformer_reduce.model.pos_embedding_3: copying a param with shape torch.Size([1, 351, 256]) from checkpoint, the shape in current model is torch.Size([1, 27, 256]).

[SonNguyen2510]

You need to retrain the model using 27 frames yourself first then using that model for your purpose.

[Vegetebird]

@SonNguyen2510 is right

[parkjin0903]

Thank you !