FocusDPO: Dynamic Preference Optimization for Multi-Subject Personalized Image Generation via Adaptive Focus

• URL: http://arxiv.org/abs/2509.01181v1
• Date: Mon, 01 Sep 2025 07:06:36 GMT
• Title: FocusDPO: Dynamic Preference Optimization for Multi-Subject Personalized Image Generation via Adaptive Focus
• Abstract summary: Multi-subject personalized image generation aims to synthesize customized images containing multiple specified subjects without requiring test-time optimization.<n>We present FocusDPO, a framework that adaptively identifies focus regions based on dynamic semantic correspondence and supervision image complexity.
• Score: 10.615833390806486
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-subject personalized image generation aims to synthesize customized images containing multiple specified subjects without requiring test-time optimization. However, achieving fine-grained independent control over multiple subjects remains challenging due to difficulties in preserving subject fidelity and preventing cross-subject attribute leakage. We present FocusDPO, a framework that adaptively identifies focus regions based on dynamic semantic correspondence and supervision image complexity. During training, our method progressively adjusts these focal areas across noise timesteps, implementing a weighted strategy that rewards information-rich patches while penalizing regions with low prediction confidence. The framework dynamically adjusts focus allocation during the DPO process according to the semantic complexity of reference images and establishes robust correspondence mappings between generated and reference subjects. Extensive experiments demonstrate that our method substantially enhances the performance of existing pre-trained personalized generation models, achieving state-of-the-art results on both single-subject and multi-subject personalized image synthesis benchmarks. Our method effectively mitigates attribute leakage while preserving superior subject fidelity across diverse generation scenarios, advancing the frontier of controllable multi-subject image synthesis.

Related papers

• Training-Free Object-Background Compositional T2I via Dynamic Spatial Guidance and Multi-Path Pruning [63.21130437262217]
  Existing text-to-image diffusion models treat the background as a passive and under-optimized byproduct.<n>We propose a training-free framework that restructures diffusion sampling to explicitly account for foreground-background interactions.
  arXiv  Detail & Related papers  (2026-04-10T19:25:24Z)
• Multimodal Large Language Models for Multi-Subject In-Context Image Generation [56.20395856287325]
  Music is the first MLLM specifically designed for textbfMUlti-textbfSubject textbfIn-textbfContext image generation.<n>We enhance the model's understanding of multi-subject semantic relationships through a vision chain-of-thought mechanism.<n>By incorporating complex subject images during training, we improve the model's capacity for chained reasoning.
  arXiv  Detail & Related papers  (2026-04-08T15:37:42Z)
• PSR: Scaling Multi-Subject Personalized Image Generation with Pairwise Subject-Consistency Rewards [86.1965460124838]
  We propose a scalable multi-subject data generation pipeline.<n>We first enable single-subject personalization models to acquire knowledge of multi-image and multi-subject scenarios.<n>To enhance both subject consistency and text controllability, we design a set of Pairwise Subject-Consistency Rewards.
  arXiv  Detail & Related papers  (2025-12-01T03:25:49Z)
• Plug-and-Play Multi-Concept Adaptive Blending for High-Fidelity Text-to-Image Synthesis [0.0]
  We introduce plug-and-play multi-concept blending for high-fidelity text-to-image (T2I) generation.<n>Our method leverages guided appearance attention to faithfully reflect the intended appearance of each personalized concept.<n>We also present a mask-guided noise mixing strategy that preserves the integrity of non-personalized regions.
  arXiv  Detail & Related papers  (2025-11-18T12:25:47Z)
• SpotDiff: Spotting and Disentangling Interference in Feature Space for Subject-Preserving Image Generation [6.116573441311417]
  SpotDiff is a novel learning-based method that extracts subject-specific features by spotting and disentangling interference.<n>To enable principled training, we introduce SpotDiff10k, a curated dataset with consistent pose and background variations.<n>Experiments demonstrate that SpotDiff achieves more robust subject preservation and controllable editing than prior methods, while attaining competitive performance with only 10k training samples.
  arXiv  Detail & Related papers  (2025-10-07T18:01:55Z)
• MultiCrafter: High-Fidelity Multi-Subject Generation via Spatially Disentangled Attention and Identity-Aware Reinforcement Learning [28.841076643572933]
  Multi-subject image generation aims to synthesize user-provided subjects in a single image.<n>Existing methods are limited by their reliance on simple reconstruction-based objectives.<n>We propose MultiCrafter, a framework that ensures high-fidelity, preference-aligned generation.
  arXiv  Detail & Related papers  (2025-09-26T06:41:43Z)
• DRC: Enhancing Personalized Image Generation via Disentangled Representation Composition [69.10628479553709]
  We introduce DRC, a novel personalized image generation framework that enhances Large Multimodal Models (LMMs)<n> DRC explicitly extracts user style preferences and semantic intentions from history images and the reference image, respectively.<n>It involves two critical learning stages: 1) Disentanglement learning, which employs a dual-tower disentangler to explicitly separate style and semantic features, optimized via a reconstruction-driven paradigm with difficulty-aware importance sampling; and 2) Personalized modeling, which applies semantic-preserving augmentations to effectively adapt the disentangled representations for robust personalized generation.
  arXiv  Detail & Related papers  (2025-04-24T08:10:10Z)
• DyMO: Training-Free Diffusion Model Alignment with Dynamic Multi-Objective Scheduling [14.621456944266802]
  We propose a training-free alignment method, DyMO, for aligning the generated images and human preferences during inference.<n>Apart from text-aware human preference scores, we introduce a semantic alignment objective for enhancing the semantic alignment in the early stages of diffusion.<n>Experiments with diverse pre-trained diffusion models and metrics demonstrate the effectiveness and robustness of the proposed method.
  arXiv  Detail & Related papers  (2024-12-01T10:32:47Z)
• TALE: Training-free Cross-domain Image Composition via Adaptive Latent Manipulation and Energy-guided Optimization [59.412236435627094]
  TALE is a training-free framework harnessing the generative capabilities of text-to-image diffusion models. We equip TALE with two mechanisms dubbed Adaptive Latent Manipulation and Energy-guided Latent Optimization. Our experiments demonstrate that TALE surpasses prior baselines and attains state-of-the-art performance in image-guided composition.
  arXiv  Detail & Related papers  (2024-08-07T08:52:21Z)
• Continual Panoptic Perception: Towards Multi-modal Incremental Interpretation of Remote Sensing Images [16.0258685984844]
  Continual learning (CL) breaks off the one-way training manner and enables a model to adapt to new data, semantics and tasks continuously. We propose a unified continual learning model that leverages multi-task joint learning covering pixel-level classification, instance-level segmentation and image-level perception.
  arXiv  Detail & Related papers  (2024-07-19T12:22:32Z)
• Coherent and Multi-modality Image Inpainting via Latent Space Optimization [61.99406669027195]
  PILOT (intextbfPainting vtextbfIa textbfLatent textbfOptextbfTimization) is an optimization approach grounded on a novel textitsemantic centralization and textitbackground preservation loss. Our method searches latent spaces capable of generating inpainted regions that exhibit high fidelity to user-provided prompts while maintaining coherence with the background.
  arXiv  Detail & Related papers  (2024-07-10T19:58:04Z)
• MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance [5.452759083801634]
  This research introduces the MS-Diffusion framework for layout-guided zero-shot image personalization with multi-subjects.<n>The proposed multi-subject cross-attention orchestrates inter-subject compositions while preserving the control of texts.
  arXiv  Detail & Related papers  (2024-06-11T12:32:53Z)
• Taming Encoder for Zero Fine-tuning Image Customization with Text-to-Image Diffusion Models [55.04969603431266]
  This paper proposes a method for generating images of customized objects specified by users. The method is based on a general framework that bypasses the lengthy optimization required by previous approaches. We demonstrate through experiments that our proposed method is able to synthesize images with compelling output quality, appearance diversity, and object fidelity.
  arXiv  Detail & Related papers  (2023-04-05T17:59:32Z)
• Auto-regressive Image Synthesis with Integrated Quantization [55.51231796778219]
  This paper presents a versatile framework for conditional image generation. It incorporates the inductive bias of CNNs and powerful sequence modeling of auto-regression. Our method achieves superior diverse image generation performance as compared with the state-of-the-art.
  arXiv  Detail & Related papers  (2022-07-21T22:19:17Z)
• DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation [56.514462874501675]
  We propose a dynamic sparse attention based Transformer model to achieve fine-level matching with favorable efficiency. The heart of our approach is a novel dynamic-attention unit, dedicated to covering the variation on the optimal number of tokens one position should focus on. Experiments on three applications, pose-guided person image generation, edge-based face synthesis, and undistorted image style transfer, demonstrate that DynaST achieves superior performance in local details.
  arXiv  Detail & Related papers  (2022-07-13T11:12:03Z)

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.