paper-with-me

홈 › Papers

LAION-SG: An Enhanced Large-Scale Dataset for Training Complex Image-Text Models with Structural Annotations

2024-12-11 · Zejian Li, Chenye Meng, Yize Li, Ling Yang, Shengyuan Zhang, Jiarui Ma, Jiayi Li, Guang Yang, Changyuan Yang, Zhiyuan Yang, Jinxiong Chang, Lingyun Sun

Recent advances in text-to-image (T2I) generation have shown remarkable success in producing high-quality images from text. However, existing T2I models show decayed performance in compositional image generation involving multiple objects and intricate relationships. We attribute this problem to limitations in existing datasets of image-text pairs, which lack precise inter-object relationship annotations with prompts only. To address this problem, we construct LAION-SG, a large-scale dataset with high-quality structural annotations of scene graphs (SG), which precisely describe attributes and relationships of multiple objects, effectively representing the semantic structure in complex scenes. Based on LAION-SG, we train a new foundation model SDXL-SG to incorporate structural annotation information into the generation process. Extensive experiments show advanced models trained on our LAION-SG boast significant performance improvements in complex scene generation over models on existing datasets. We also introduce CompSG-Bench, a benchmark that evaluates models on compositional image generation, establishing a new standard for this domain. Our annotations with the associated processing code, the foundation model and the benchmark protocol are publicly available at https://github.com/mengcye/LAION-SG.

📄 PDF Abstract BibTeX arXiv:2412.08580

Code (2)

mengcye/LAION-SG 공식 구현 pytorch
yangling0818/sgdiff pytorch

Tasks

AttributeImage GenerationScene Generation

Similar Papers 제목 키워드 기반

Unmasking LAION-5B: Age, Gender, Race, and Emotion Biases in Large-Scale Image Datasets

2026-06-22 · Iris Dominguez-Catena, Daniel Paternain, Mikel Galar arxiv

Large-scale image-text datasets, such as LAION-5B, are foundational to modern AI systems, yet their vast scale and uncurated nature raise significant concerns about demographic and stereotypical biases. This study presen…

LAION-5B: An open large-scale dataset for training next generation image-text models

2022-10-16 · NeurIPS 2022 Datasets and Benchmarks 2022 6 · Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon 외

Groundbreaking language-vision architectures like CLIP and DALL-E proved the utility of training on large amounts of noisy image-text data, without relying on expensive accurate labels used in standard vision unimodal su…

Image GenerationPreference Mappingzero-shot-classificationZero-Shot Learning

LAION-BVD: A 10-Million-Hour Open Video Dataset for Multimodal Pre-training

2026-08-25 · Andreas Hochlehnert, Marianna Nezhurina, Mehdi Cherti, Andrej Radonjic 외 hf

We present LAION-BVD, a large-scale open video dataset for multimodal learning, which contains 1.3B platform-specific video URLs collected from CommonCrawl. From these, we download 80M videos with a total duration of 10 …

Text Retrieval

Structured Captions Improve Prompt Adherence in Text-to-Image Models (Re-LAION-Caption 19M)

2025-07-07 · Nicholas Merchant, Haitz Sáez de Ocáriz Borde, Andrei Cristian Popescu, Carlos Garcia Jurado Suarez arxiv

We argue that generative text-to-image models often struggle with prompt adherence due to the noisy and unstructured nature of large-scale datasets like LAION-5B. This forces users to rely heavily on prompt engineering t…

Visual Question AnsweringPrompt Engineering

Effective pruning of web-scale datasets based on complexity of concept clusters

2024-01-09 · Amro Abbas, Evgenia Rusak, Kushal Tirumala, Wieland Brendel 외

Utilizing massive web-scale datasets has led to unprecedented performance gains in machine learning models, but also imposes outlandish compute requirements for their training. In order to improve training and data effic…