Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2022
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2202.01284 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916044904333312 |
|---|---|
| author | Jakob, Wenzel Speierer, Sébastien Roussel, Nicolas Vicini, Delio |
| author_facet | Jakob, Wenzel Speierer, Sébastien Roussel, Nicolas Vicini, Delio |
| contents | Dr$.$Jit is a new just-in-time compiler for physically based rendering and its derivative. Dr$.$Jit expedites research on these topics in two ways: first, it traces high-level simulation code (e.g., written in Python) and aggressively simplifies and specializes the resulting program representation, producing data-parallel kernels with state-of-the-art performance on CPUs and GPUs.
Second, it simplifies the development of differentiable rendering algorithms. Efficient methods in this area turn the derivative of a simulation into a simulation of the derivative. Dr$.$Jit provides fine-grained control over the process of automatic differentiation to help with this transformation.
Specialization is particularly helpful in the context of differentiation, since large parts of the simulation ultimately do not influence the computed gradients. Dr$.$Jit tracks data dependencies globally to find and remove redundant computation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2202_01284 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Dr.Jit: A Just-In-Time Compiler for Differentiable Rendering Jakob, Wenzel Speierer, Sébastien Roussel, Nicolas Vicini, Delio Graphics Dr$.$Jit is a new just-in-time compiler for physically based rendering and its derivative. Dr$.$Jit expedites research on these topics in two ways: first, it traces high-level simulation code (e.g., written in Python) and aggressively simplifies and specializes the resulting program representation, producing data-parallel kernels with state-of-the-art performance on CPUs and GPUs. Second, it simplifies the development of differentiable rendering algorithms. Efficient methods in this area turn the derivative of a simulation into a simulation of the derivative. Dr$.$Jit provides fine-grained control over the process of automatic differentiation to help with this transformation. Specialization is particularly helpful in the context of differentiation, since large parts of the simulation ultimately do not influence the computed gradients. Dr$.$Jit tracks data dependencies globally to find and remove redundant computation. |
| title | Dr.Jit: A Just-In-Time Compiler for Differentiable Rendering |
| topic | Graphics |
| url | https://arxiv.org/abs/2202.01284 |