domain randomization

This study addresses the low physical fidelity of generative video models—manifested as artifacts such as jittering and interpenetration—by proposing a physics-aware enhancement method grounded in synthetic video. Methodologically, it employs a differentiable rendering pipeline to generate physically consistent synthetic videos, establishes a physics-perceptive data filtering mechanism, and introduces cross-domain feature alignment coupled with adversarial physical consistency regularization—enabling physics realism transfer without differentiable simulation or explicit physical modeling. This work provides the first empirical evidence that synthetic video can substantially improve physical fidelity in video generation. Evaluated on three physics-sensitive tasks—rigid-body collisions, fluid motion, and pendulum dynamics—the approach reduces physical violation rates significantly, achieving an average 37.2% improvement in physical plausibility, validated jointly by user studies and automated physical violation detection.

The scarcity of real-world data severely hinders the widespread adoption of subsymbolic AI. To address this challenge, this work proposes a unified reference framework based on digital twins to systematically design and analyze simulation-based synthetic data generation methods for AI training. By integrating digital twin technology, high-fidelity simulation, and synthetic data generation, the framework delineates core components, advantages, and key challenges, offering a methodological foundation for producing high-quality, reproducible training data. This study not only fills the critical gap in the lack of systematic guidance for synthetic data generation but also provides a scalable and reusable technical pathway to mitigate reliance on real-world data.

This work addresses the significant appearance gap between synthetic and real images—commonly referred to as the sim2real appearance gap—that limits the applicability of synthetic data in real-world vision tasks. The authors propose a hybrid augmentation framework that, for the first time, integrates the geometric and material generation capabilities of the diffusion model FLUX.2-4B Klein with the distribution-matching strengths of the image-to-image translation model REGEN. This combination enhances visual realism while preserving semantic consistency. Experimental results demonstrate that the proposed approach substantially narrows the sim2real appearance gap, outperforming individual models in overall quality, with REGEN contributing notably superior photorealism.

This work addresses the critical gap in understanding whether synthetic images are truly interchangeable with real ones in model training and the absence of systematic evaluation frameworks to ensure their safe and effective use. The study systematically quantifies discrepancies between synthetic and real images across three dimensions: high-dimensional feature distributions, low-level statistical properties in color space, and model training dynamics. Building on these insights, the authors propose a pre-evaluation metric for synthetic data of unknown quality and a safety-aware data fusion strategy for training. Experiments demonstrate that carefully calibrated mixing ratios and integration methods of synthetic and real data can substantially enhance model performance and robustness, thereby offering both theoretical grounding and practical guidance for the reliable deployment of synthetic data in machine learning pipelines.

This paper addresses the scarcity and high acquisition cost of real-world data in computer vision by proposing a novel paradigm for synthetic image generation and evaluation using Unreal Engine 4. Methodologically, it introduces a synthetic image quality assessment framework anchored on pre-trained models as auditors and employs a synthetic–real hybrid training strategy, validated on VGG16 and MobileNetV3-small. Key contributions include: (1) achieving comparable performance to conventional training with 50–70% real data using only 10% real images; (2) reducing the test–train accuracy gap to 1–2% by incorporating >60% synthetic images, thereby significantly mitigating domain shift; and (3) halving the error rate in purely synthetic training through the addition of <10% real samples—enabling strong trade-offs among data privacy, cost efficiency, and experimental reproducibility.

Controllable human video generation is hindered by the scarcity of real-world data, particularly for rare identities and complex motion scenarios. This work proposes a unified diffusion-based framework that systematically investigates, for the first time, the synergistic mechanisms between synthetic and real data in human-centric video generation. It reveals their complementary roles and introduces an efficient synthetic sample selection strategy to enhance training. The proposed approach significantly improves motion realism, temporal coherence, and identity fidelity in generated videos, establishing a new paradigm for building data-efficient and generalizable controllable video generation models.

To address domain shift between synthetic and real-world images and high annotation costs in robotic vision tasks, this paper proposes an automated training dataset generation pipeline tailored for robotic environments. Methodologically, it introduces a novel two-pass rendering framework based on 3D Gaussian Splatting, integrating proxy-mesh shadow mapping with splatting-based image synthesis to achieve physically plausible shadow and highlight modeling. Concurrently, it generates pixel-accurate segmentation masks compatible with mainstream detectors such as YOLO. By training on a hybrid dataset comprising a small set of real images and large-scale, high-fidelity synthetic data, the pipeline significantly improves object detection and instance segmentation accuracy. Experiments demonstrate that the approach effectively bridges the domain gap while maintaining high rendering efficiency, offering a scalable, efficient paradigm for building robust robotic vision models.

This study systematically evaluates the suitability and effectiveness of synthetic data across three canonical scenarios: data sharing, model training augmentation, and variance reduction in statistical estimation. By integrating formal modeling, theoretical analysis of generative models, and empirical case studies, the work presents the first comprehensive taxonomy of synthetic data applications and delineates their boundaries of applicability. The research elucidates both the potential and fundamental limitations of synthetic data in enhancing privacy preservation, model performance, and statistical stability. It further demonstrates that many existing or proposed use cases are misaligned with the intrinsic properties of synthetic data, thereby providing decision-makers with a principled theoretical framework to assess whether synthetic data is appropriate for addressing specific data availability challenges.

This work proposes a GAN-inspired privacy-preserving synthetic data generation method that avoids direct access to original data during training. Instead, it leverages fuzz testing to produce candidate samples and iteratively refines them through a discriminator-guided feedback loop combined with statistical distribution constraints to approximate the original data distribution. By innovatively integrating fuzz testing, adversarial discrimination, and indirect constraint mechanisms, the approach achieves strong privacy guarantees—effectively resisting membership inference and data reconstruction attacks—while preserving high data utility. Extensive experiments on four benchmark datasets demonstrate that the proposed method strikes a superior balance between privacy protection and data fidelity compared to existing techniques.

This work addresses the lack of reproducible and quantifiable evaluation benchmarks in existing digital twin generation methods, which often rely on subjective qualitative comparisons. To this end, the paper proposes a synthetic image generation framework based on high-fidelity 3D models and programmable camera poses, enabling systematic quantitative assessment of reconstruction results under known ground-truth parameters. The approach introduces, for the first time, a programmable virtual environment coupled with a ground-truth parameter reference mechanism, integrating procedural trajectory generation, photorealistic rendering, and feature-point triangulation-based reconstruction. This framework establishes the first benchmark for digital twin evaluation that supports reproducible and objective comparisons, significantly enhancing the consistency and scientific rigor of assessments across different generation strategies.