Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis
Jian Han, Jinlai Liu, Yi Jiang, Bin Yan, Yuqi Zhang, Zehuan Yuan, Bingyue Peng, Xiaobing Liu
2024-12-06
Summary
This paper talks about Infinity, a new model that generates high-resolution images from text descriptions using advanced techniques that improve both speed and image quality.
What's the problem?
Generating high-quality images from text prompts can be slow and resource-intensive with existing models. Many of these models also struggle to create detailed images and often require extensive adjustments to work effectively.
What's the solution?
The authors introduced Infinity, which uses a unique method called Bitwise Visual AutoRegressive Modeling. This approach allows the model to predict image details more effectively by using an infinite vocabulary for image features and a self-correction mechanism. Infinity can generate high-quality images at a resolution of 1024x1024 in just 0.8 seconds, making it significantly faster than previous models. It also outperforms other leading models in quality and detail.
Why it matters?
This research is important because it sets a new standard for generating images from text, making the process faster and more efficient. By improving the way images are created, Infinity could enhance applications in fields like graphic design, video game development, and virtual reality, where high-quality visuals are essential.
Abstract
We present Infinity, a Bitwise Visual AutoRegressive Modeling capable of generating high-resolution, photorealistic images following language instruction. Infinity redefines visual autoregressive model under a bitwise token prediction framework with an infinite-vocabulary tokenizer & classifier and bitwise self-correction mechanism, remarkably improving the generation capacity and details. By theoretically scaling the tokenizer vocabulary size to infinity and concurrently scaling the transformer size, our method significantly unleashes powerful scaling capabilities compared to vanilla VAR. Infinity sets a new record for autoregressive text-to-image models, outperforming top-tier diffusion models like SD3-Medium and SDXL. Notably, Infinity surpasses SD3-Medium by improving the GenEval benchmark score from 0.62 to 0.73 and the ImageReward benchmark score from 0.87 to 0.96, achieving a win rate of 66%. Without extra optimization, Infinity generates a high-quality 1024x1024 image in 0.8 seconds, making it 2.6x faster than SD3-Medium and establishing it as the fastest text-to-image model. Models and codes will be released to promote further exploration of Infinity for visual generation and unified tokenizer modeling.