Question: How to get ray indices from the Nerfacto model inside the get_outputs() function

[stanathong]

Hello, I was wondering if anyone know how to get ray indices from the Nerfacto model.

In the Instant NGP model, you get ray indices when calling proposal_sampler() inside the get_outputs() function, which is quite convenient to know which ray indices the samples are from.

nerfstudio/nerfstudio/models/instant_ngp.py

Lines 173 to 178 in 189328e

	def get_outputs(self, ray_bundle: RayBundle):
	assert self.field is not None
	num_rays = len(ray_bundle)
	with torch.no_grad():
	ray_samples, ray_indices = self.sampler(

In the Nerfacto model, you don't get the ray indices. Is there a way to retrieve ray indices inside the get_outputs() function as that of the Instant NGP model, or somehow derived them from somewhere?

nerfstudio/nerfstudio/models/nerfacto.py

Lines 298 to 303 in 189328e

	def get_outputs(self, ray_bundle: RayBundle):
	# apply the camera optimizer pose tweaks
	if self.training:
	self.camera_optimizer.apply_to_raybundle(ray_bundle)
	ray_samples: RaySamples
	ray_samples, weights_list, ray_samples_list = self.proposal_sampler(ray_bundle, density_fns=self.density_fns)

Thank you very much.