Norbert Kirchgessner

Olivia Zumsteg, Jannis Widmer, Yann Bourdé et al.

Accurate estimation of wheat spike volume is important for yield component analysis and stress resilience assessment, yet field-based measurement remains challenging. Active 3D sensing methods such as Light Detection and Ranging (LiDAR) or time-of-flight (ToF) are sensitive to plant motion or poorly suited to outdoor conditions, while 3D reconstructions are computationally expensive. Direct 2D image processing would offer computational advantages, but image-based models lack explicit geometric information. We therefore propose a hybrid 2D-3D approach with knowledge distillation during training while enabling efficient image-only inference. First, we train a rigid-invariant point cloud network using distance-based histogram features to obtain pose-robust geometric representations. We then combine the 3D model with a proposed multi-view image-based regulated Transformer (RT) in an ensemble architecture. Finally, we distill the ensemble knowledge into a purely image-based student model using either feature-based or label-based distillation. The two distilled RTs reduce the mean absolute error (MAE) from 654.31 mm$^3$ of the non-distilled RT to 639.93 mm$^3$ and 644.62 mm$^3$, and increase correlation from 0.76 to 0.77 and 0.82, respectively. At the same time, inference time is reduced from 160 ms to 1.4 ms per spike. Distillation further mitigates volume-dependent bias and reshapes the latent representation of the image model toward a geometry-aware shape. Our results demonstrate that 3D-informed training of a 2D Transformer allows for scalable and efficient spike volume estimation for high-throughput field phenotyping.

Daiwei Zhang, Joaquin Gajardo, Tomislav Medic et al.

Automated extraction of plant morphological traits is crucial for supporting crop breeding and agricultural management through high-throughput field phenotyping (HTFP). Solutions based on multi-view RGB images are attractive due to their scalability and affordability, enabling volumetric measurements that 2D approaches cannot directly capture. While advanced methods like Neural Radiance Fields (NeRFs) have shown promise, their application has been limited to counting or extracting traits from only a few plants or organs. Furthermore, accurately measuring complex structures like individual wheat heads-essential for studying crop yields-remains particularly challenging due to occlusions and the dense arrangement of crop canopies in field conditions. The recent development of 3D Gaussian Splatting (3DGS) offers a promising alternative for HTFP due to its high-quality reconstructions and explicit point-based representation. In this paper, we present Wheat3DGS, a novel approach that leverages 3DGS and the Segment Anything Model (SAM) for precise 3D instance segmentation and morphological measurement of hundreds of wheat heads automatically, representing the first application of 3DGS to HTFP. We validate the accuracy of wheat head extraction against high-resolution laser scan data, obtaining per-instance mean absolute percentage errors of 15.1%, 18.3%, and 40.2% for length, width, and volume. We provide additional comparisons to NeRF-based approaches and traditional Muti-View Stereo (MVS), demonstrating superior results. Our approach enables rapid, non-destructive measurements of key yield-related traits at scale, with significant implications for accelerating crop breeding and improving our understanding of wheat development.

Olivia Zumsteg, Nico Graf, Aaron Haeusler et al.

Estimating three-dimensional morphological traits from two-dimensional RGB images presents inherent challenges due to the loss of depth information, projection distortions, and occlusions under field conditions. In this work, we explore multiple approaches for non-destructive volume estimation of wheat spikes, using RGB image sequences and structured-light 3D scans as ground truth references. Due to the complex geometry of the spikes, we propose a neural network approach for volume estimation in 2D images, employing a transfer learning pipeline that combines DINOv2, a self-supervised Vision Transformer, with a unidirectional Long Short-Term Memory (LSTM) network. By using deep supervision, the model is able to learn more robust intermediate representations, which enhances its generalisation ability across varying evaluation sequences. We benchmark our model against two conventional baselines: a 2D area-based projection and a geometric reconstruction using axis-aligned cross-sections. Our deep supervised model achieves a mean absolute percentage error (MAPE) of 6.46% on six-view indoor images, outperforming the area (9.36%) and geometric (13.98%) baselines. Fine-tuning the model on field-based single-image data enables domain adaptation, yielding a MAPE of 10.82%. We demonstrate that object shape significantly impacts volume prediction accuracy, with irregular geometries such as wheat spikes posing greater challenges for geometric methods compared to our deep learning approach.