Christoph Heindl
Results
122
comments of
Christoph Heindl
@yihao19 yes that's possible and demonstrated [here](https://github.com/cheind/pytorch-blender/tree/develop/examples/densityopt). It requires back-prop through non-differentiable render function of Blender, which I documented [there](https://github.com/cheind/pytorch-blender/blob/develop/examples/densityopt/etc/blendtorch_stochopt.pdf)
yes, that would be required, but it probably isn't. You might use one of the gradient estimator 'tricks' to avoid that - reparametrization or score function gradient estimation are keywords....