InstanceGen: Image Generation with Instance-level Instructions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sella, Etai, Kleiman, Yanir, Averbuch-Elor, Hadar
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910949975261184
author Sella, Etai
Kleiman, Yanir
Averbuch-Elor, Hadar
author_facet Sella, Etai
Kleiman, Yanir
Averbuch-Elor, Hadar
contents Despite rapid advancements in the capabilities of generative models, pretrained text-to-image models still struggle in capturing the semantics conveyed by complex prompts that compound multiple objects and instance-level attributes. Consequently, we are witnessing growing interests in integrating additional structural constraints, typically in the form of coarse bounding boxes, to better guide the generation process in such challenging cases. In this work, we take the idea of structural guidance a step further by making the observation that contemporary image generation models can directly provide a plausible fine-grained structural initialization. We propose a technique that couples this image-based structural guidance with LLM-based instance-level instructions, yielding output images that adhere to all parts of the text prompt, including object counts, instance-level attributes, and spatial relations between instances.
format Preprint
id arxiv_https___arxiv_org_abs_2505_05678
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle InstanceGen: Image Generation with Instance-level Instructions
Sella, Etai
Kleiman, Yanir
Averbuch-Elor, Hadar
Computer Vision and Pattern Recognition
Despite rapid advancements in the capabilities of generative models, pretrained text-to-image models still struggle in capturing the semantics conveyed by complex prompts that compound multiple objects and instance-level attributes. Consequently, we are witnessing growing interests in integrating additional structural constraints, typically in the form of coarse bounding boxes, to better guide the generation process in such challenging cases. In this work, we take the idea of structural guidance a step further by making the observation that contemporary image generation models can directly provide a plausible fine-grained structural initialization. We propose a technique that couples this image-based structural guidance with LLM-based instance-level instructions, yielding output images that adhere to all parts of the text prompt, including object counts, instance-level attributes, and spatial relations between instances.
title InstanceGen: Image Generation with Instance-level Instructions
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.05678