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 introduces Infinity, a new method for generating high-resolution, realistic images based on language instructions using advanced modeling techniques.
What's the problem?
Current image generation methods often struggle with producing high-quality images quickly and efficiently, particularly when they need to follow specific text prompts. Many existing models require significant computational power and can be slow, which makes them less practical for real-time applications.
What's the solution?
Infinity addresses these challenges by using a Bitwise Visual AutoRegressive Modeling approach. It features an infinite-vocabulary tokenizer and a bitwise self-correction mechanism that allows the model to generate images with greater detail and accuracy. By scaling both the tokenizer and the model size, Infinity can produce high-quality images much faster—around 0.8 seconds for a 1024x1024 image—making it significantly quicker than other models. The researchers demonstrated that Infinity outperforms leading models in various benchmarks, achieving higher scores in image quality assessments.
Why it matters?
This research is important because it sets a new standard for image generation technology, making it faster and more efficient while maintaining high quality. This advancement can benefit many fields, including gaming, film, and virtual reality, where quick and realistic image generation is 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.