# Beyond Global Realism: Virtual Try-On Evaluation and Optimization with Dimension-wise Garment Fidelity Assessment

arXiv: https://arxiv.org/abs/2608.29804 · HTML: https://mlsanity.com/paper/2608.29804/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 47 references that are in the map.
- First posted on arXiv 30 Aug 2026.

## Builds on (most cited first, 24 of 24)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [VITON-HD: High-Resolution Virtual Try-On via Misalignment-Aware Normalization](https://mlsanity.com/paper/2103.16874/) (2021)
- [VITON: An Image-based Virtual Try-on Network](https://mlsanity.com/paper/1711.08447/) (2017)
- [Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective](https://mlsanity.com/paper/2303.14968/) (2023)
- [Taming the Power of Diffusion Models for High-Quality Virtual Try-On with Appearance Flow](https://mlsanity.com/paper/2308.06101/) (2023)
- [Aligning Large Language Models through Synthetic Feedback](https://mlsanity.com/paper/2305.13735/) (2023)
- [RewardDance: Reward Scaling in Visual Generation](https://mlsanity.com/paper/2509.08826/) (2025)
- [R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning](https://mlsanity.com/paper/2505.02835/) (2025)
- [EditScore: Unlocking Online RL for Image Editing via High-Fidelity Reward Modeling](https://mlsanity.com/paper/2509.23909/) (2025)
- [ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding](https://mlsanity.com/paper/2507.14533/) (2025)
- [Texture-Preserving Diffusion Models for High-Fidelity Virtual Try-On](https://mlsanity.com/paper/2404.01089/) (2024)
- [Street TryOn: Learning In-the-Wild Virtual Try-On from Unpaired Person Images](https://mlsanity.com/paper/2311.16094/) (2023)
- [DiffSim: Taming Diffusion Models for Evaluating Visual Similarity](https://mlsanity.com/paper/2412.14580/) (2024)
- [PICTURE: PhotorealistIC virtual Try-on from UnconstRained dEsigns](https://mlsanity.com/paper/2312.04534/) (2023)
- [Fine-grained Image Quality Assessment for Perceptual Image Restoration](https://mlsanity.com/paper/2508.14475/) (2025)
- [Fine-grained Image Aesthetic Assessment: Learning Discriminative Scores from Relative Ranks](https://mlsanity.com/paper/2603.03907/) (2026)
- [PosterReward: Unlocking Accurate Evaluation for High-Quality Graphic Design Generation](https://mlsanity.com/paper/2603.29855/) (2026)

## Abstract (arXiv)

Virtual try-on (VTON) requires not only realistic generation but also faithful preservation of garment characteristics. However, existing evaluation metrics such as PSNR, SSIM, KID and FID struggle to measure the consistency between the generated and reference garments, particularly in capturing the multi-dimensional characteristics of garment fidelity. To address this, we propose DAT: a Dimension-wise Assessment framework for virtual Try-on, which decomposes garment consistency into seven interpretable dimensions: silhouette, color, neckline and sleeve shape, major decoration and structure, material texture, fine-detail fidelity, and logo preservation, each formulated as a discrete attribute-level prediction task. To train this specialized assessment model, we adopt a two-stage learning paradigm comprising large-scale weak supervision on 50K samples, followed by refinement on 10K higher-quality annotations obtained via multi-model voting. Furthermore, we employ weighted cross-entropy loss to mitigate the severe label imbalance inherent across evaluation dimensions. Beyond its role as an evaluation framework, the assessment model can be integrated into reinforcement learning optimization of Qwen-Image-Edit for VTON, where dimension-wise rewards are adaptively aggregated to emphasize under-optimized aspects during training. Experimental results show that our method (8B parameters) achieves state-of-the-art performance in terms of balanced accuracy, SROCC, and PLCC, outperforming strong proprietary models such as Gemini-3.1, Qwen3.7-plus, and GPT-5.5, while also serving as an effective optimization signal for reward-guided VTON generation