Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Rendering

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Youwang, Kim, Oh, Tae-Hyun, Pons-Moll, Gerard
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917659755413504
author Youwang, Kim
Oh, Tae-Hyun
Pons-Moll, Gerard
author_facet Youwang, Kim
Oh, Tae-Hyun
Pons-Moll, Gerard
contents We present Paint-it, a text-driven high-fidelity texture map synthesis method for 3D meshes via neural re-parameterized texture optimization. Paint-it synthesizes texture maps from a text description by synthesis-through-optimization, exploiting the Score-Distillation Sampling (SDS). We observe that directly applying SDS yields undesirable texture quality due to its noisy gradients. We reveal the importance of texture parameterization when using SDS. Specifically, we propose Deep Convolutional Physically-Based Rendering (DC-PBR) parameterization, which re-parameterizes the physically-based rendering (PBR) texture maps with randomly initialized convolution-based neural kernels, instead of a standard pixel-based parameterization. We show that DC-PBR inherently schedules the optimization curriculum according to texture frequency and naturally filters out the noisy signals from SDS. In experiments, Paint-it obtains remarkable quality PBR texture maps within 15 min., given only a text description. We demonstrate the generalizability and practicality of Paint-it by synthesizing high-quality texture maps for large-scale mesh datasets and showing test-time applications such as relighting and material control using a popular graphics engine. Project page: https://kim-youwang.github.io/paint-it
format Preprint
id arxiv_https___arxiv_org_abs_2312_11360
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Rendering
Youwang, Kim
Oh, Tae-Hyun
Pons-Moll, Gerard
Computer Vision and Pattern Recognition
Artificial Intelligence
Graphics
We present Paint-it, a text-driven high-fidelity texture map synthesis method for 3D meshes via neural re-parameterized texture optimization. Paint-it synthesizes texture maps from a text description by synthesis-through-optimization, exploiting the Score-Distillation Sampling (SDS). We observe that directly applying SDS yields undesirable texture quality due to its noisy gradients. We reveal the importance of texture parameterization when using SDS. Specifically, we propose Deep Convolutional Physically-Based Rendering (DC-PBR) parameterization, which re-parameterizes the physically-based rendering (PBR) texture maps with randomly initialized convolution-based neural kernels, instead of a standard pixel-based parameterization. We show that DC-PBR inherently schedules the optimization curriculum according to texture frequency and naturally filters out the noisy signals from SDS. In experiments, Paint-it obtains remarkable quality PBR texture maps within 15 min., given only a text description. We demonstrate the generalizability and practicality of Paint-it by synthesizing high-quality texture maps for large-scale mesh datasets and showing test-time applications such as relighting and material control using a popular graphics engine. Project page: https://kim-youwang.github.io/paint-it
title Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Rendering
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Graphics
url https://arxiv.org/abs/2312.11360