Salvato in:
Dettagli Bibliografici
Autori principali: Tasar, Onur, Chadebec, Clément, Aubin, Benjamin
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2412.11972
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916526171357184
author Tasar, Onur
Chadebec, Clément
Aubin, Benjamin
author_facet Tasar, Onur
Chadebec, Clément
Aubin, Benjamin
contents Realistic shadow generation is a critical component for high-quality image compositing and visual effects, yet existing methods suffer from certain limitations: Physics-based approaches require a 3D scene geometry, which is often unavailable, while learning-based techniques struggle with control and visual artifacts. We introduce a novel method for fast, controllable, and background-free shadow generation for 2D object images. We create a large synthetic dataset using a 3D rendering engine to train a diffusion model for controllable shadow generation, generating shadow maps for diverse light source parameters. Through extensive ablation studies, we find that rectified flow objective achieves high-quality results with just a single sampling step enabling real-time applications. Furthermore, our experiments demonstrate that the model generalizes well to real-world images. To facilitate further research in evaluating quality and controllability in shadow generation, we release a new public benchmark containing a diverse set of object images and shadow maps in various settings. The project page is available at https://gojasper.github.io/controllable-shadow-generation-project/
format Preprint
id arxiv_https___arxiv_org_abs_2412_11972
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Controllable Shadow Generation with Single-Step Diffusion Models from Synthetic Data
Tasar, Onur
Chadebec, Clément
Aubin, Benjamin
Computer Vision and Pattern Recognition
Realistic shadow generation is a critical component for high-quality image compositing and visual effects, yet existing methods suffer from certain limitations: Physics-based approaches require a 3D scene geometry, which is often unavailable, while learning-based techniques struggle with control and visual artifacts. We introduce a novel method for fast, controllable, and background-free shadow generation for 2D object images. We create a large synthetic dataset using a 3D rendering engine to train a diffusion model for controllable shadow generation, generating shadow maps for diverse light source parameters. Through extensive ablation studies, we find that rectified flow objective achieves high-quality results with just a single sampling step enabling real-time applications. Furthermore, our experiments demonstrate that the model generalizes well to real-world images. To facilitate further research in evaluating quality and controllability in shadow generation, we release a new public benchmark containing a diverse set of object images and shadow maps in various settings. The project page is available at https://gojasper.github.io/controllable-shadow-generation-project/
title Controllable Shadow Generation with Single-Step Diffusion Models from Synthetic Data
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.11972