Summary of Text2avatar: Text to 3d Human Avatar Generation with Codebook-driven Body Controllable Attribute, by Chaoqun Gong et al.
Text2Avatar: Text to 3D Human Avatar Generation with Codebook-Driven Body Controllable Attribute
by Chaoqun Gong, Yuqin Dai, Ronghui Li, Achun Bao, Jun Li, Jian Yang, Yachao Zhang, Xiu Li
First submitted to arxiv on: 1 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Text2Avatar model generates realistic 3D human avatars based on text prompts, leveraging a discrete codebook to establish a connection between text and avatars. This approach enables the disentanglement of features, addressing challenges in multi-attribute controllable and realistic avatar generation. By utilizing pre-trained unconditional 3D human avatar generation models to obtain pseudo data, Text2Avatar can generate realistic-style 3D avatars from coupled textual data, outperforming existing methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Text2Avatar is a new way to make 3D human characters from text. This helps with making character modeling cheaper and faster. The problem is that it’s hard to control the features of these characters, like what they look like or how they move. Text2Avatar uses a special book of codes to connect text to characters, letting us change one feature without affecting others. It also uses fake data from other models to learn how to make realistic characters. This works better than other ways to do this. |
