Image Understanding Makes for A Good Tokenizer for Image Generation

• URL: http://arxiv.org/abs/2411.04406v1
• Date: Thu, 07 Nov 2024 03:55:23 GMT
• Title: Image Understanding Makes for A Good Tokenizer for Image Generation
• Authors: Luting Wang, Yang Zhao, Zijian Zhang, Jiashi Feng, Si Liu, Bingyi Kang,
• Abstract summary: We introduce a token-based IG framework, which relies on effective tokenizers to project images into token sequences. We show that tokenizers with strong IU capabilities achieve superior IG performance across a variety of metrics, datasets, tasks, and proposal networks.
• Score: 62.875788091204626
• License:
• Abstract: Abstract Modern image generation (IG) models have been shown to capture rich semantics valuable for image understanding (IU) tasks. However, the potential of IU models to improve IG performance remains uncharted. We address this issue using a token-based IG framework, which relies on effective tokenizers to project images into token sequences. Currently, pixel reconstruction (e.g., VQGAN) dominates the training objective for image tokenizers. In contrast, our approach adopts the feature reconstruction objective, where tokenizers are trained by distilling knowledge from pretrained IU encoders. Comprehensive comparisons indicate that tokenizers with strong IU capabilities achieve superior IG performance across a variety of metrics, datasets, tasks, and proposal networks. Notably, VQ-KD CLIP achieves $4.10$ FID on ImageNet-1k (IN-1k). Visualization suggests that the superiority of VQ-KD can be partly attributed to the rich semantics within the VQ-KD codebook. We further introduce a straightforward pipeline to directly transform IU encoders into tokenizers, demonstrating exceptional effectiveness for IG tasks. These discoveries may energize further exploration into image tokenizer research and inspire the community to reassess the relationship between IU and IG. The code is released at https://github.com/magic-research/vector_quantization.

Related papers

• Learning Graph Quantized Tokenizers for Transformers [28.79505338383552]
  Graph Transformers (GTs) have emerged as a leading model in deep learning, outperforming Graph Neural Networks (GNNs) in various graph learning tasks. We introduce GQT (textbfGraph textbfQuantized textbfTokenizer), which decouples tokenizer training from Transformer training by leveraging graph self-supervised learning. By combining the GQT with token modulation, a Transformer encoder achieves state-of-the-art performance on 16 out of 18 benchmarks, including large-scale homophilic and heterophilic datasets.
  arXiv  Detail & Related papers  (2024-10-17T17:38:24Z)
• SGC-VQGAN: Towards Complex Scene Representation via Semantic Guided Clustering Codebook [9.993066868670283]
  We introduce SGC-VQGAN through Semantic Online Clustering method to enhance token semantics through Consistent Semantic Learning. Our approach constructs a temporospatially consistent semantic codebook, addressing issues of codebook collapse and imbalanced token semantics.
  arXiv  Detail & Related papers  (2024-09-09T23:12:43Z)
• Rejuvenating image-GPT as Strong Visual Representation Learners [28.77567067712619]
  This paper enhances image-GPT, one of the pioneering works that introduce autoregressive pretraining to predict the next pixels. We shift the prediction target from raw pixels to semantic tokens, enabling a higher-level understanding of visual content. Experiments showcase that D-iGPT excels as a strong learner of visual representations.
  arXiv  Detail & Related papers  (2023-12-04T18:59:20Z)
• ADDP: Learning General Representations for Image Recognition and Generation with Alternating Denoising Diffusion Process [94.41510903676837]
  We propose an Alternating Denoising Diffusion Process (ADDP) that integrates two spaces within a single representation learning framework. In each denoising step, our method first decodes pixels from previous VQ tokens, then generates new VQ tokens from the decoded pixels. The learned representations can be used to generate diverse high-fidelity images and also demonstrate excellent transfer performance on recognition tasks.
  arXiv  Detail & Related papers  (2023-06-08T17:59:32Z)
• Unpaired Image Captioning by Image-level Weakly-Supervised Visual Concept Recognition [83.93422034664184]
  Unpaired image captioning (UIC) is to describe images without using image-caption pairs in the training phase. Most existing studies use off-the-shelf algorithms to obtain the visual concepts. We propose a novel approach to achieve cost-effective UIC using image-level labels.
  arXiv  Detail & Related papers  (2022-03-07T08:02:23Z)
• Semantic-Aware Generation for Self-Supervised Visual Representation Learning [116.5814634936371]
  We advocate for Semantic-aware Generation (SaGe) to facilitate richer semantics rather than details to be preserved in the generated image. SaGe complements the target network with view-specific features and thus alleviates the semantic degradation brought by intensive data augmentations. We execute SaGe on ImageNet-1K and evaluate the pre-trained models on five downstream tasks including nearest neighbor test, linear classification, and fine-scaled image recognition.
  arXiv  Detail & Related papers  (2021-11-25T16:46:13Z)
• Vector-quantized Image Modeling with Improved VQGAN [93.8443646643864]
  We propose a Vector-quantized Image Modeling approach that involves pretraining a Transformer to predict image tokens autoregressively. We first propose multiple improvements over vanilla VQGAN from architecture to codebook learning, yielding better efficiency and reconstruction fidelity. When trained on ImageNet at 256x256 resolution, we achieve Inception Score (IS) of 175.1 and Frechet Inception Distance (FID) of 4.17, a dramatic improvement over the vanilla VQGAN.
  arXiv  Detail & Related papers  (2021-10-09T18:36:00Z)
• Attention-Driven Dynamic Graph Convolutional Network for Multi-Label Image Recognition [53.17837649440601]
  We propose an Attention-Driven Dynamic Graph Convolutional Network (ADD-GCN) to dynamically generate a specific graph for each image. Experiments on public multi-label benchmarks demonstrate the effectiveness of our method.
  arXiv  Detail & Related papers  (2020-12-05T10:10:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.