The paper introduces a multibranch ensemble head that pairs low- and high-capacity branches for Vision Foundation Models, sampling one branch during training and averaging both at inference. This lightweight design adds negligible parameters and inference overhead while improving accuracy-efficiency trade-offs on data-scarce tasks. Applied to cucumber phenological stage classification, the DINOv2 ViT-S with the multibranch head achieves a Macro F1 of 0.978 and Cohen's Kappa of 0.971, matching larger models but with only 12.2 GFlops and 22.3 M parameters, enabling deployment on smartphones and edge devices.
By Kostiantyn Khabarlak
A100 is a hybrid CNN‑Transformer vision foundation model trained on 3 million unlabeled agricultural images using self‑supervised techniques (DINO, iBOT, KoLeo). It achieves superior crop classification (95.37 % with a frozen backbone) and phenotyping accuracy, and after fine‑tuning reaches 99.90 % on 32‑class crop identification and 99.02 % on 44‑class disease detection. With only 1 % of labeled data, A100 attains 98.01 % plant classification accuracy, outperforming existing baselines and other agricultural foundation models by significant margins.
By Mayur Pratim Das, Ajai Dubey, Aksshay Mathew P, Sana Afreen, Rishi Kumar Soni, Deepak Meena, Pratibha Kumari, Mukesh Saini
arXiv:2608.21254v1 Announce Type: cross
Abstract: Accurate agricultural weed detection in real-world field conditions is essential for precision agriculture, enabling targeted intervention and reduci...
By Nikhilesh Prabhakar, Pranuthi Tenali, Wilfredo Abudeye Fernandez, Shekhar Borah, Athresh Karanam, Erik Blasch, Prabha Sundaravadivel, Sriraam Natarajan
The paper introduces a dual‑stream Swin ResNet hybrid architecture that combines CNN and Transformer streams for cross‑grade banana variety classification. It employs KL‑divergence‑based self‑distillation to align the probabilistic outputs of the two streams, achieving a mean accuracy of 81.25% on a dataset of 31,678 images across three varieties and three quality grades. Experiments show the hybrid model outperforms both CNN and Transformer baselines, improving feature transferability, attention localization, and optimization stability.
By Khush Kataruka, Mohammad Muzammila, Zahid Akhtar, Raghavendra Ramachandra, Praveen Kumar Chandaliya
arXiv:2603.27519v4 Announce Type: replace
Abstract: Image-based plant phenotyping depends on dense structural understanding of crops, yet pixel-level annotation remains expensive across species, orga...
By Shuai Xiang, James Burridge, Shouyang Liu, Hao Lu, Tokihiro Fukatsu, Yinqiang Zheng, Wei Guo
The paper introduces AGS-PlantSeg, a few‑shot 3D plant organ segmentation method that uses the frozen Utonia foundation model and Adaptive Granularity Selection (AGS) to dynamically choose optimal spatial granularity for each plant. By extracting tailored geometric features for a lightweight MLP head, AGS-PlantSeg achieves superior cross‑species generalization, reaching an average mIoU of 88.9% and outperforming fixed‑granularity baselines by 2.5 points across PLANesT‑3D, Pheno4D, and Crops3D datasets. The approach requires minimal annotated data yet competes with fully supervised, plant‑specific architectures.
arXiv:2605. 05627v2 Announce Type: replace-cross Abstract: Sustainable forest management relies on precise species composition mapping, yet traditional ground surveys are labour-intensive and geographically constrained.
By Gabriel Jeanson, David-Alexandre Duclos, William Larriv\'ee-Hardy, No\'e Cochet, Mat\v{e}j Boxan, Anthony Desch\^enes, Fran\c{c}ois Pomerleau, Philippe Gigu\`ere
AT‑ViT is a dual‑branch Vision Transformer that processes both raw herbarium scans and their segmentation masks through a multi‑scale, multi‑view cross‑attention fusion. It uses a mask‑guided patch weighting scheme to emphasize plant regions and suppress background artifacts, thereby encouraging plant‑centric representations. In trait classification tasks such as leaf base shape and thorns, AT‑ViT consistently outperforms baselines, improves spatial attention grounding (IoU_p +15.66 to +18.03 pp, IoU_b –27.92 to –31.02 pp), and shows greater robustness to synthetic background perturbations, surpassing ResNet101 by up to +32.32 accuracy points and CrossViT by up to +5.07 points.
whyItMatters":"The model addresses shortcut learning caused by background cues in herbarium images, leading to more accurate and interpretable plant trait recognition."
By Amani Sedrat, Takieddine Chehhat, Youcef Sklab, Hanane Ariouat, Abderrazak Sebaa, Eric Chenin, Jean-Daniel Zucker, Edi Profiti
arXiv:2608.21454v1 Announce Type: new
Abstract: The same fruit appears in a bunch, unpicked, peeled, bagged in plastic, or sliced on a dish, so automated fruit classification in the wild (AFCW) must...
By Subhankar Chattoraj, Sawon Pratiher, Samiran Das, Hubert Konik
LeafTrackNet is a deep learning framework that combines a YOLOv10-based leaf detector with a MobileNetV3-based embedding network to track individual leaves over time. The authors introduce CanolaTrack, a large benchmark dataset of 5,704 RGB images with 31,840 annotated leaf instances from 184 canola plants. When evaluated without prior fine‑tuning, LeafTrackNet outperforms existing methods on CanolaTrack, KOMATSUNA, and MSU‑PID datasets, achieving HOTA scores of 88.03, 87.33, and 74.20 respectively.
By Shanghua Liu, Majharulislam Babor, Christoph Verduyn, Breght Vandenberghe, Bruno Betoni Parodi, Cornelia Weltzien, Marina M. -C. H\"ohne
CoAtNet-DeepMoE is a lightweight Convolution‑Attention hybrid architecture that incorporates a DeepSeek Mixture‑of‑Experts to reduce parameters while maintaining high accuracy for tomato disease classification. The model achieves state‑of‑the‑art performance on Kaggle and PlantVillage datasets, reporting 99.80% accuracy on Kaggle and 99.83% accuracy on PlantVillage, all with only 2.47 million parameters. The source code will be released on GitHub.
By Md Nadim Mahamood, Md Arif Shahriar, Md Shafi Ud Doula, Kamrul Hasan
The study benchmarks Vision Transformers (ViTs) against convolutional neural networks (CNNs) for fine‑grained orchid genus identification in New Guinea’s species‑rich, data‑poor flora. Using a two‑stage system that first predicts genus and then retrieves similar species images, the authors fine‑tuned four pretrained backbones on 16,701 photographs from 120 genera and 1,350 species. The self‑supervised ViT DINOv2 achieved the highest genus accuracy (macro top‑1 66.9 %) and outperformed both CNNs and a domain‑matched pretrained ViT, demonstrating strong species retrieval and open‑set detection capabilities.
By Reza Saputra, Diah Harnoni Apriyanti, Andr\'e Schuiteman, Kurt Metzger, Ashley Field, Katharina Nargar, William Edwards