PhenoBench: A Large Dataset and Benchmarks for Semantic Image Interpretation in the Agricultural Domain.

The production of food, feed, fiber, and fuel is a key task of agriculture, which has to cope with many challenges in the upcoming decades, e.g., a higher demand, climate change, lack of workers, and the availability of arable land. Vision systems can support making better and more sustainable field management decisions, but also support the breeding of new crop varieties by allowing temporally dense and reproducible measurements. Recently, agricultural robotics got an increasing interest in the vision and robotics communities since it is a promising avenue for coping with the aforementioned lack of workers and enabling more sustainable production. While large datasets and benchmarks in other domains are readily available and enable significant progress, agricultural datasets and benchmarks are comparably rare. We present an annotated dataset and benchmarks for the semantic interpretation of real agricultural fields. Our dataset recorded with a UAV provides high-quality, pixel-wise annotations of crops and weeds, but also crop leaf instances at the same time. Furthermore, we provide benchmarks for various tasks on a hidden test set comprised of different fields: known fields covered by the training data and a completely unseen field. Our dataset, benchmarks, and code are available at https://www.phenobench.org.

VDBFusion: Flexible and Efficient TSDF Integration of Range Sensor Data.

Mapping is a crucial task in robotics and a fundamental building block of most mobile systems deployed in the real world. Robots use different environment representations depending on their task and sensor setup. This paper showcases a practical approach to volumetric surface reconstruction based on truncated signed distance functions, also called TSDFs. We revisit the basics of this mapping technique and offer an approach for building effective and efficient real-world mapping systems. In contrast to most state-of-the-art SLAM and mapping approaches, we are making no assumptions on the size of the environment nor the employed range sensor. Unlike most other approaches, we introduce an effective system that works in multiple domains using different sensors. To achieve this, we build upon the Academy-Award-winning OpenVDB library used in filmmaking to realize an effective 3D map representation. Based on this, our proposed system is flexible and highly effective and, in the end, capable of integrating point clouds from a 64-beam LiDAR sensor at 20 frames per second using a single-core CPU. Along with this publication comes an easy-to-use C++ and Python library to quickly and efficiently solve volumetric mapping problems with TSDFs.