Computer vision training dataset generation for robotic environments using Gaussian splatting

📅 2025-12-15
📈 Citations: 0
Influential: 0
📄 PDF

career value

225K/year
🤖 AI Summary
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.

Technology Category

Application Category

📝 Abstract
This paper introduces a novel pipeline for generating large-scale, highly realistic, and automatically labeled datasets for computer vision tasks in robotic environments. Our approach addresses the critical challenges of the domain gap between synthetic and real-world imagery and the time-consuming bottleneck of manual annotation. We leverage 3D Gaussian Splatting (3DGS) to create photorealistic representations of the operational environment and objects. These assets are then used in a game engine where physics simulations create natural arrangements. A novel, two-pass rendering technique combines the realism of splats with a shadow map generated from proxy meshes. This map is then algorithmically composited with the image to add both physically plausible shadows and subtle highlights, significantly enhancing realism. Pixel-perfect segmentation masks are generated automatically and formatted for direct use with object detection models like YOLO. Our experiments show that a hybrid training strategy, combining a small set of real images with a large volume of our synthetic data, yields the best detection and segmentation performance, confirming this as an optimal strategy for efficiently achieving robust and accurate models.
Problem

Research questions and friction points this paper is trying to address.

Generates realistic synthetic datasets for robotic vision
Bridges domain gap between synthetic and real-world imagery
Automates labeling to eliminate manual annotation bottlenecks
Innovation

Methods, ideas, or system contributions that make the work stand out.

Leveraging 3D Gaussian Splatting for photorealistic environment representations
Using two-pass rendering with shadow maps to enhance synthetic image realism
Automatically generating pixel-perfect segmentation masks for direct model training
🔎 Similar Papers
No similar papers found.
P
Patryk Niżeniec
Institute of Engineering and Technology, Faculty of Physics, Astronomy and Informatics, Nicolaus Copernicus University in Toruń, 87-100, Toruń, Poland.
M
Marcin Iwanowski
Institute of Engineering and Technology, Faculty of Physics, Astronomy and Informatics, Nicolaus Copernicus University in Toruń, 87-100, Toruń, Poland.