Search papers, labs, and topics across Lattice.
Oxygen-TryOn introduces a fashion-native foundation model designed specifically for any-item virtual try-on, overcoming limitations of existing systems that focus on single garment categories. By utilizing a dedicated data engine and a three-stage training approach, including continued pre-training, supervised fine-tuning, and reinforcement learning, it synthesizes photorealistic images of subjects wearing diverse fashion items while maintaining identity and item appearance. The model outperforms existing benchmarks in both single- and multi-item try-on scenarios, achieving state-of-the-art results in consistency and realism compared to leading proprietary and open-source systems.
Oxygen-TryOn achieves unprecedented realism in virtual try-on by synthesizing images across diverse fashion categories, outperforming both proprietary and open-source models.
We present Oxygen-TryOn, a unified foundation model for any-item virtual try-on. Rather than repurposing a general-purpose image editor, Oxygen-TryOn is fashion-native, built for try-on through a dedicated data engine and try-on-specific training. Given one or more reference items (clean product shots or in-the-wild worn-on photos) and a single target subject image, it synthesizes a photorealistic image of the subject wearing the items across virtually any fashion category. Prior systems handle a single garment category in a studio setting, and recent multi-reference methods remain garment-centric; in contrast, Oxygen-TryOn supports diverse items and scenarios, including full- and half-body views, a variable number of references, and free multi-item composition, while faithfully preserving both subject identity and item appearance. Instead of mask-based inpainting, we reformulate try-on as a multi-reference, understanding-driven generation task. We build a data engine that collects, manufactures, annotates, and filters high-quality try-on data at scale, and design a three-stage recipe of continued pre-training (CPT), supervised fine-tuning (SFT), and reinforcement learning (RL). The RL stage uses a hybrid reward combining an in-house try-on reward model with a proprietary, rubric-guided general-purpose model, jointly supervising fine-grained consistency and instruction-level quality. It also follows general editing instructions (e.g., pose changes) in the same pass. Across public benchmarks and our in-house Oxygen-TryOn Bench, it achieves state-of-the-art consistency and realism on single-item try-on and leads on multi-item try-on, matching or surpassing both leading proprietary systems (Nano Banana Pro, GPT-Image-2, Seedream5 Lite) and open-source models (FLUX.2).
