More Than Generation: Unifying Generation and Depth Estimation via Text-to-Image Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Hongkai, Liang, Dingkang, Du, Mingyang, Zhou, Xin, Bai, Xiang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918173393027072
author Lin, Hongkai
Liang, Dingkang
Du, Mingyang
Zhou, Xin
Bai, Xiang
author_facet Lin, Hongkai
Liang, Dingkang
Du, Mingyang
Zhou, Xin
Bai, Xiang
contents Generative depth estimation methods leverage the rich visual priors stored in pre-trained text-to-image diffusion models, demonstrating astonishing zero-shot capability. However, parameter updates during training lead to catastrophic degradation in the image generation capability of the pre-trained model. We introduce MERGE, a unified model for image generation and depth estimation, starting from a fixed pre-trained text-to-image model. MERGE demonstrates that the pre-trained text-to-image model can do more than image generation, but also expand to depth estimation effortlessly. Specifically, MERGE introduces a play-and-plug framework that enables seamless switching between image generation and depth estimation modes through simple and pluggable converters. Meanwhile, we propose a Group Reuse Mechanism to encourage parameter reuse and improve the utilization of the additional learnable parameters. MERGE unleashes the powerful depth estimation capability of the pre-trained text-to-image model while preserving its original image generation ability. Compared to other unified models for image generation and depth estimation, MERGE achieves state-of-the-art performance across multiple depth estimation benchmarks. The code will be made available at https://github.com/H-EmbodVis/MERGE
format Preprint
id arxiv_https___arxiv_org_abs_2510_23574
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle More Than Generation: Unifying Generation and Depth Estimation via Text-to-Image Diffusion Models
Lin, Hongkai
Liang, Dingkang
Du, Mingyang
Zhou, Xin
Bai, Xiang
Computer Vision and Pattern Recognition
Generative depth estimation methods leverage the rich visual priors stored in pre-trained text-to-image diffusion models, demonstrating astonishing zero-shot capability. However, parameter updates during training lead to catastrophic degradation in the image generation capability of the pre-trained model. We introduce MERGE, a unified model for image generation and depth estimation, starting from a fixed pre-trained text-to-image model. MERGE demonstrates that the pre-trained text-to-image model can do more than image generation, but also expand to depth estimation effortlessly. Specifically, MERGE introduces a play-and-plug framework that enables seamless switching between image generation and depth estimation modes through simple and pluggable converters. Meanwhile, we propose a Group Reuse Mechanism to encourage parameter reuse and improve the utilization of the additional learnable parameters. MERGE unleashes the powerful depth estimation capability of the pre-trained text-to-image model while preserving its original image generation ability. Compared to other unified models for image generation and depth estimation, MERGE achieves state-of-the-art performance across multiple depth estimation benchmarks. The code will be made available at https://github.com/H-EmbodVis/MERGE
title More Than Generation: Unifying Generation and Depth Estimation via Text-to-Image Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.23574