OverLay++: Dense-Overlap Layout-to-Image Generation Dataset
cs.CV
Submitted: 2026-10-06
Updated: 2026-10-06
Comments: Accepted at NeurIPS 2026, Evaluations & Datasets Track. Project website: https://mlpc-ucsd.github.io/OverLayPP . Dataset: https://huggingface.co/datasets/mlpcucsd/OverLayPP
Code: https://github.com/kakaobrain/coyo-dataset
Project page: https://mlpc-ucsd.github.io/OverLayPP
License: http://creativecommons.org/licenses/by/4.0/
The gist: Layout-to-Image generation has made substantial progress in spatial and object-level control.
Terminology
Abstract
Layout-to-Image generation has made substantial progress in spatial and object-level control. However, existing methods still struggle with complex scenes containing many overlapping and interacting objects. We argue that training data is a particular bottleneck: existing datasets lack examples with dense, complex object interactions. To address this gap, we introduce OverLay++, a large-scale Layout-to-Image dataset with structurally complex scenes. OverLay++ contains approximately 500K images with an average of 6.6 objects per image, exceeding existing datasets by 1.67 times in annotation density. Beyond annotation density, OverLay++ provides rich semantic detail with object captions over six times longer than in current datasets. Our dataset generation pipeline is simple and produces dense, overlapping object annotations with rich per-object captions. Across multiple benchmarks, state-of-the-art Layout-to-Image methods trained on the OverLay++ dataset show consistent improvement and faster convergence, demonstrating the importance of dense, overlap-aware, and caption-rich supervision for controllable image generation.
Sources
- OminiControl: Minimal and Universal Control for Diffusion Transformer
- DICEPTION: A Generalist Diffusion Model for Visual Perceptual Tasks
- 3DIS: Depth-Driven Decoupled Instance Synthesis for Text-to-Image Generation
- 3DIS-FLUX: simple and efficient multi-instance generation with DiT rendering
- CreatiLayout: Siamese Multimodal Diffusion Transformer for Creative Layout-to-Image Generation
- EliGen: Entity-Level Controlled Image Generation with Regional Attention
- MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis
- Denoising Diffusion Implicit Models
- Classifier-Free Diffusion Guidance
- Hierarchical Text-Conditional Image Generation with CLIP Latents
- Lay-Your-Scene: Natural Scene Layout Generation with Diffusion Transformers
- LAION-5B: An open large-scale dataset for training next generation image-text models
- OverLayBench: A Benchmark for Layout-to-Image Generation with Dense Overlaps
- Kosmos-2: Grounding Multimodal Large Language Models to the World
- Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- LLM Blueprint: Enabling Text-to-Image Generation with Complex and Detailed Prompts
- Training-free Regional Prompting for Diffusion Transformers
- HiCo: Hierarchical Controllable Diffusion Model for Layout-to-image Generation
- Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models