1Texas A&M University,
2Astroblox AI,
3The University of Texas at Austin
*Equal contribution, The order is random.
Text: “A cute Corgi”
Text-to-3D generation using Stable DreamFusion.
Avoiding Janus problem in text-to-3D generation with Stable DreamFusion + Perp-Neg .
The proposed Perp-Neg alleviates the Janus problem in text-to-3D generations, and achieves view generation in a more precise way in the text-to-image generation task. Beyond that, it can also be used to generate natural 2D images while eliminating undersired attributes from the negative text descriptions.
Abstract
Although text-to-image diffusion models have made significant strides in generating images from text, they are sometimes more inclined to generate images like the data on which the model was trained rather than the provided text.
This limitation has hindered their usage in both 2D and 3D applications.
To address this problem, we explored the use of negative prompts but found that the current implementation fails to produce desired results, particularly when there is an overlap between the main and
