Morten Stigaard Laursen

Alfred T. Christiansen, Andreas H. Højrup, Morten K. Stephansen et al.

Training neural networks for tasks such as 3D point cloud semantic segmentation demands extensive datasets, yet obtaining and annotating real-world point clouds is costly and labor-intensive. This work aims to introduce a novel pipeline for generating realistic synthetic data, by leveraging 3D Gaussian Splatting (3DGS) and Gaussian Opacity Fields (GOF) to generate 3D assets of multiple different agricultural vehicles instead of using generic models. These assets are placed in a simulated environment, where the point clouds are generated using a simulated LiDAR. This is a flexible approach that allows changing the LiDAR specifications without incurring additional costs. We evaluated the impact of synthetic data on segmentation models such as PointNet++, Point Transformer V3, and OACNN, by training and validating the models only on synthetic data. Remarkably, the PTv3 model had an mIoU of 91.35\%, a noteworthy result given that the model had neither been trained nor validated on any real data. Further studies even suggested that in certain scenarios the models trained only on synthetically generated data performed better than models trained on real-world data. Finally, experiments demonstrated that the models can generalize across semantic classes, enabling accurate predictions on mesh models they were never trained on.

Filip Slezak, Magnus K. Gjerde, Joakim B. Haurum et al.

In this paper, we introduce a 3D Gaussian Splatting (3DGS)-based pipeline for stereo dataset generation, offering an efficient alternative to Neural Radiance Fields (NeRF)-based methods. To obtain useful geometry estimates, we explore utilizing the reconstructed geometry from the explicit 3D representations as well as depth estimates from the FoundationStereo model in an expert knowledge transfer setup. We find that when fine-tuning stereo models on 3DGS-generated datasets, we demonstrate competitive performance in zero-shot generalization benchmarks. When using the reconstructed geometry directly, we observe that it is often noisy and contains artifacts, which propagate noise to the trained model. In contrast, we find that the disparity estimates from FoundationStereo are cleaner and consequently result in a better performance on the zero-shot generalization benchmarks. Our method highlights the potential for low-cost, high-fidelity dataset creation and fast fine-tuning for deep stereo models. Moreover, we also reveal that while the latest Gaussian Splatting based methods have achieved superior performance on established benchmarks, their robustness falls short in challenging in-the-wild settings warranting further exploration.

Martin Peter Christiansen, Morten Stigaard Laursen, Rasmus Nyholm Jørgensen et al.

Mapping field environments into point clouds using a 3D LIDAR has the ability to become a new approach for online estimation of crop biomass in the field. The estimation of crop biomass in agriculture is expected to be closely correlated to canopy heights. The work presented in this paper contributes to the mapping and textual analysis of agricultural fields. Crop and environmental state information can be used to tailor treatments to the specific site. This paper presents the current results with our ground vehicle LiDAR mapping systems for broad acre crop fields. The proposed vehicle system and method facilitates LiDAR recordings in an experimental winter wheat field. LiDAR data are combined with data from Global Navigation Satellite System (GNSS) and Inertial Measurement Unit (IMU) sensors to conduct environment mapping for point clouds. The sensory data from the vehicle are recorded, mapped, and analyzed using the functionalities of the Robot Operating System (ROS) and the Point Cloud Library (PCL). In this experiment winter wheat (Triticum aestivum L.) in field plots, was mapped using 3D point clouds with a point density on the centimeter level. The purpose of the experiment was to create 3D LiDAR point-clouds of the field plots enabling canopy volume and textural analysis to discriminate different crop treatments. Estimated crop volumes ranging from 3500-6200 (m3) per hectare are correlated to the manually collected samples of cut biomass extracted from the experimental field.

Mikkel Fly Kragh, Peter Christiansen, Morten Stigaard Laursen et al.

In this paper, we present a novel multi-modal dataset for obstacle detection in agriculture. The dataset comprises approximately 2 hours of raw sensor data from a tractor-mounted sensor system in a grass mowing scenario in Denmark, October 2016. Sensing modalities include stereo camera, thermal camera, web camera, 360-degree camera, lidar, and radar, while precise localization is available from fused IMU and GNSS. Both static and moving obstacles are present including humans, mannequin dolls, rocks, barrels, buildings, vehicles, and vegetation. All obstacles have ground truth object labels and geographic coordinates.