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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

3D generation has raised great attention in recent years. With the success oftext-to-image diffusion models, the 2D-lifting technique becomes a promisingroute to controllable 3D generation. However, these methods tend to presentinconsistent geometry, which is also known as the Janus problem. We observethat the problem is caused mainly by two aspects, i.e., viewpoint bias in 2Ddiffusion models and overfitting of the optimization objective. To address it,we propose a two-stage 2D-lifting framework, namely DreamControl, whichoptimizes coarse NeRF scenes as 3D self-prior and then generates fine-grainedobjects with control-based score distillation. Specifically, adaptive viewpointsampling and boundary integrity metric are proposed to ensure the consistencyof generated priors. The priors are then regarded as input conditions tomaintain reasonable geometries, in which conditional LoRA and weighted scoreare further proposed to optimize detailed textures. DreamControl can generatehigh-quality 3D content in terms of both geometry consistency and texturefidelity. Moreover, our control-based optimization guidance is applicable tomore downstream tasks, including user-guided generation and 3D animation. Theproject page is available at https://github.com/tyhuang0428/DreamControl.

DreamControl: Control-Based Text-to-3D Generation with 3D Self-Prior