A Hybrid Approach for Closing the Sim2real Appearance Gap in Game Engine Synthetic Datasets
For computer vision researchers using synthetic data, this work offers a practical method to improve photorealism, though the gains are incremental as it combines existing techniques.
The paper investigates the sim2real appearance gap in synthetic datasets from game engines, finding that a hybrid approach combining a diffusion model (FLUX.2-4B Klein) with an image-to-image translation model (REGEN) achieves better visual realism than either model alone while maintaining semantic consistency.
Video game engines have been an important source for generating large volumes of visual synthetic datasets for training and evaluating computer vision algorithms that are to be deployed in the real world. While the visual fidelity of modern game engines has been significantly improved with technologies such as ray-tracing, a notable sim2real appearance gap between the synthetic and the real-world images still remains, which limits the utilization of synthetic datasets in real-world applications. In this letter, we investigate the ability of a state-of-the-art image generation and editing diffusion model (FLUX.2-4B Klein) to enhance the photorealism of synthetic datasets and compare its performance against a traditional image-to-image translation model (REGEN). Furthermore, we propose a hybrid approach that combines the strong geometry and material transformations of diffusion-based methods with the distribution-matching capabilities of image-to-image translation techniques. Through experiments, it is demonstrated that REGEN outperforms FLUX.2-4B Klein and that by combining both FLUX.2-4B Klein and REGEN models, better visual realism can be achieved compared to using each model individually, while maintaining semantic consistency. The code is available at: https://github.com/stefanos50/Hybrid-Sim2Real