English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

3D generation has rapidly accelerated in the past decade owing to theprogress in the field of generative modeling. Score Distillation Sampling (SDS)based rendering has improved 3D asset generation to a great extent. Further,the recent work of Denoising Diffusion Policy Optimization (DDPO) demonstratesthat the diffusion process is compatible with policy gradient methods and hasbeen demonstrated to improve the 2D diffusion models using an aesthetic scoringfunction. We first show that this aesthetic scorer acts as a strong guide for avariety of SDS-based methods and demonstrates its effectiveness in text-to-3Dsynthesis. Further, we leverage the DDPO approach to improve the quality of the3D rendering obtained from 2D diffusion models. Our approach, DDPO3D, employsthe policy gradient method in tandem with aesthetic scoring. To the best of ourknowledge, this is the first method that extends policy gradient methods to 3Dscore-based rendering and shows improvement across SDS-based methods such asDreamGaussian, which are currently driving research in text-to-3D synthesis.Our approach is compatible with score distillation-based methods, which wouldfacilitate the integration of diverse reward functions into the generativeprocess. Our project page can be accessed via https://ddpo3d.github.io.

RL Dreams: Policy Gradient Optimization for Score Distillation based 3D Generation