Closed yzslab closed 1 year ago
Do you mean the optimized NeRF parameters for each individual scene? No, we have no plan to release those models.
Note that NeRF models do not have a traditional "training" step --- when NeRF is applied to a set of images, its weights are randomly initialized and then optimized with those images using gradient descent. There is no "pretraining".
Do you mean the optimized NeRF parameters for each individual scene? No, we have no plan to release those models.
Note that NeRF models do not have a traditional "training" step --- when NeRF is applied to a set of images, its weights are randomly initialized and then optimized with those images using gradient descent. There is no "pretraining".
OK, thanks for your reply.
Hi, Seems like it need to take a lot of time to replicate the paper results if training from scratch. It will be better if you can provide some pretrained weights. Thanks.