pith. sign in

arxiv: 2607.01753 · v1 · pith:FM6W5SQAnew · submitted 2026-07-02 · 💻 cs.CV · q-bio.QM

The Turning Point of 3D Plant Phenotyping: 3D Foundation Models Enable Minute-to-Second Cross-Crop Reconstruction and Beyond

Hanyue Jia , Wei Zhou , Wenbo Zhou , Yanan Li , Hao Lu , Tingting Wu This is my paper

Pith reviewed 2026-07-03 16:32 UTC · model grok-4.3

classification 💻 cs.CV q-bio.QM
keywords 3D plant phenotyping3D foundation modelsGaussian splattingfeed-forward reconstructioncross-crop datasetsmartphone imagingphenotypic measurementfew-view reconstruction
0
0 comments X

The pith

3D foundation models reduce plant phenotyping reconstruction from minutes to seconds across crops.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that 3D Foundation Models can replace the slow, multi-step traditional pipeline for 3D plant phenotyping with a faster feed-forward approach. Standard methods need extensive multi-view capture, fragile reconstruction like COLMAP, and separate steps to extract phenotypes, which becomes impractical with sparse smartphone images. The new framework uses 3DFM-based geometric recovery, geometry-constrained 3D Gaussian Splatting, iterative view synthesis, and 2D-to-3D semantic transfer to go from images to measurable organs. Experiments on 26 sequences from a new cross-crop smartphone dataset confirm the time drop while keeping reconstruction quality and phenotyping accuracy. This opens a route to high-throughput 3D phenotyping from low-cost data.

Core claim

3D Foundation Models enable a cross-crop 3D phenotyping framework that replaces COLMAP-style sparse initialization with feed-forward geometric recovery, combines it with geometry-constrained 3D Gaussian Splatting for dense reconstruction, supports few-view cases through iterative synthesis and refinement, and converts geometry to phenotypic measurements via 2D-to-3D semantic transfer, metric scale recovery, and organ separation, cutting average reconstruction time from 6.52 minutes to 1.58 seconds on 26 plant sequences while preserving quality and accuracy.

What carries the argument

3D Foundation Model feed-forward geometric recovery integrated with geometry-constrained 3D Gaussian Splatting and 2D-to-3D semantic transfer for organ instance separation and measurement.

If this is right

  • Smartphone videos become sufficient for reliable 3D plant reconstruction and measurement.
  • The same pipeline applies across diverse plant morphologies without per-crop redesign.
  • Phenotypic extraction becomes feasible at much higher throughput from low-cost captures.
  • A new cross-crop dataset with manual annotations supports further evaluation of segmentation and traits.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach could extend to other biological 3D tasks like animal morphology if the same feed-forward recovery holds.
  • Testing on even fewer views or outdoor field conditions would clarify the limits of the iterative synthesis step.
  • Combining the output with automated trait databases could create end-to-end pipelines for crop breeding.

Load-bearing premise

General 3D foundation models pre-trained on everyday scenes recover accurate geometry for many plant shapes from sparse smartphone views without plant-specific fine-tuning or loss of phenotyping accuracy.

What would settle it

Reconstruction quality or phenotyping error rates that drop sharply on a new plant species with morphology outside the tested set would show the models do not transfer without adaptation.

Figures

Figures reproduced from arXiv: 2607.01753 by Hanyue Jia, Hao Lu, Tingting Wu, Wei Zhou, Wenbo Zhou, Yanan Li.

Figure 1
Figure 1. Figure 1: Comparison between classical and 3DFM-based pipelines for 3D plant phenotyping. Classical pipelines typically proceed from image acquisition to COLMAP/SfM-based initialization, dense reconstruction, and manual or interactive trait extraction, and are limited by minute-level initialization, dense-view dependence, and manual intervention. The proposed pipeline replaces the matching-based front end with 3DFM-… view at source ↗
Figure 2
Figure 2. Figure 2: Dataset overview and acquisition protocol. Smartphone videos were acquired along closed-loop trajectories with height-adaptive shooting and frame sampling. The dataset included a robustness subset spanning multiple species and conditions, and a quantitative subset with manual annotations for segmentation and phenotypic evaluation. ments followed the official configurations of VGGT and π 3 , based on Python… view at source ↗
Figure 3
Figure 3. Figure 3: Pipeline for few-view 3D plant reconstruction, segmentation, and phenotyping. The workflow starts from sparse multi-view inputs, performs 3DFM-based initialization and Mip-Splatting-based densification, supplements effective observations through Difix3D+-assisted view refinement, and then converts the reconstructed geometry into organ-level traits through 2D-to-3D semantic transfer, scale recovery, leaf in… view at source ↗
Figure 4
Figure 4. Figure 4: Effect of Input View Count on Reconstruction and Phenotyping Per￾formance. PSNR, SSIM, and LPIPS are reported for COLMAP and the proposed π 3 + novel view synthesis pipeline under different numbers of input views. The proposed method achieves usable reconstruction at much lower view counts and approaches the dense-view performance ceiling more rapidly. the proposed workflow under low-view input, we compare… view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative comparison of plant organ segmentation results across methods. Qualitative results of PSegNet, TPointNet++, PointTransformerV3, and the proposed method on soybean, maize, and sesame samples, with GT shown for reference. As shown in [PITH_FULL_IMAGE:figures/full_fig_p025_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Phenotypic measurement accuracy across crops and growth stages. Top: cross-crop agreement between estimated and manual leaf area and inclination angle. Bottom: stage-wise leaf-area Estimated/GT ratios and signed leaf-inclination deviations, with shaded reference ranges shown for area ratios and inclination deviations, respectively. 28 [PITH_FULL_IMAGE:figures/full_fig_p028_6.png] view at source ↗
read the original abstract

3D plant phenotyping is notoriously known to be procedure-complicated and of low throughput due to the extensive multi-view imaging, the fragile 3D reconstruction pipeline, and the additional cost from reconstructed geometry to phenotypic extraction. These limitations are further amplified in low-cost data acquisition, where smartphone videos or sparsely sampled multi-view images provide limited view overlap and self-occlusion. In this work, we show that the conventional 3D plant phenotyping pipeline could be streamlined and significantly accelerated with 3D Foundation Models (3DFMs), and particularly, present one of the first cross-crop 3D phenotyping frameworks powered by 3DFMs. The framework replaces COLMAP-style sparse initialization with 3DFM-based feed-forward geometric recovery, combines geometry-constrained 3D Gaussian Splatting for dense reconstruction, enables few-view reconstruction through iterative view synthesis and refinement, and converts reconstructed geometry into measurable organs through 2D-to-3D semantic transfer, metric scale recovery, and organ instance separation. We further construct a cross-crop dataset with smartphone-based image acquisition, diverse plant morphologies, and manual annotations for segmentation and phenotypic evaluation. Experiments across 26 plant sequences show that 3D Foundation Models reduce the average reconstruction time from 6.52 minutes to 1.58 seconds while maintaining high reconstruction quality and phenotyping accuracy. These results suggest a fresh technical route for high-throughput 3D plant phenotyping, from low-cost image acquisition to fast reconstruction, perception, scale recovery, and phenotypic measurement.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces a 3D plant phenotyping pipeline that substitutes COLMAP-style sparse initialization with feed-forward geometric recovery from 3D Foundation Models (3DFMs), followed by geometry-constrained 3D Gaussian Splatting, iterative view synthesis for few-view cases, and 2D-to-3D semantic transfer for organ-level phenotyping. It constructs a new cross-crop dataset from smartphone imagery with manual annotations and reports experiments on 26 sequences claiming average reconstruction time drops from 6.52 minutes to 1.58 seconds while preserving reconstruction quality and phenotyping accuracy.

Significance. If the empirical claims hold under scrutiny, the work offers a practical route to high-throughput 3D phenotyping from low-cost, sparse smartphone captures across diverse morphologies, directly addressing throughput bottlenecks in conventional multi-view pipelines. The construction of a cross-crop dataset with segmentation and phenotypic annotations is a concrete contribution that could support future benchmarking.

major comments (2)
  1. [Abstract] Abstract: the headline claim that 3DFMs 'maintain high reconstruction quality and phenotyping accuracy' is presented without any quantitative metrics, error bars, baseline comparisons (e.g., against COLMAP or other SfM methods), dataset statistics, or method hyperparameters, rendering the central empirical result unverifiable from the provided text.
  2. [Section 3] Section 3: the pipeline performs direct substitution of COLMAP with 3DFM inference without plant-specific fine-tuning or domain adaptation; if feed-forward depth/normal estimates systematically deviate on thin stems, self-occlusions, or non-Lambertian leaf surfaces typical of the cross-crop set, the subsequent Gaussian Splatting and 2D-to-3D transfer steps cannot recover metric phenotyping accuracy, falsifying the 'maintaining high phenotyping accuracy' clause.
minor comments (1)
  1. [Abstract] The abstract and results section should explicitly state the number of views per sequence, camera intrinsics handling, and the precise phenotyping metrics (e.g., leaf area, stem length) used for accuracy evaluation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which help clarify the presentation of our empirical claims. We address each major point below and indicate revisions to the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the headline claim that 3DFMs 'maintain high reconstruction quality and phenotyping accuracy' is presented without any quantitative metrics, error bars, baseline comparisons (e.g., against COLMAP or other SfM methods), dataset statistics, or method hyperparameters, rendering the central empirical result unverifiable from the provided text.

    Authors: We agree that the abstract is currently too high-level. The full manuscript reports quantitative results (time reduction from 6.52 min to 1.58 s across 26 sequences, plus quality and accuracy metrics versus COLMAP baselines) in Sections 4 and 5, including dataset statistics. We will revise the abstract to incorporate the key numerical findings, error ranges where applicable, and explicit baseline references so the central claim is verifiable from the abstract alone. revision: yes

  2. Referee: [Section 3] Section 3: the pipeline performs direct substitution of COLMAP with 3DFM inference without plant-specific fine-tuning or domain adaptation; if feed-forward depth/normal estimates systematically deviate on thin stems, self-occlusions, or non-Lambertian leaf surfaces typical of the cross-crop set, the subsequent Gaussian Splatting and 2D-to-3D transfer steps cannot recover metric phenotyping accuracy, falsifying the 'maintaining high phenotyping accuracy' clause.

    Authors: The manuscript intentionally uses off-the-shelf 3DFMs without fine-tuning to demonstrate cross-crop generality. The cross-crop dataset explicitly includes the morphologies mentioned (thin stems, occlusions, non-Lambertian leaves). Our results in Section 5 show that the full pipeline (3DFM initialization + geometry-constrained Gaussian Splatting + 2D-to-3D transfer) preserves phenotyping accuracy relative to COLMAP baselines on this data. We will add a short paragraph in Section 3.2 discussing the robustness mechanisms and note any residual failure modes observed on the most challenging sequences. revision: partial

Circularity Check

0 steps flagged

No significant circularity; empirical results on constructed dataset

full rationale

The paper presents an empirical pipeline that substitutes 3DFM feed-forward inference for COLMAP, applies Gaussian Splatting and view synthesis, then measures reconstruction time and phenotyping accuracy on a newly constructed cross-crop smartphone dataset across 26 sequences. The headline result (6.52 min to 1.58 s) is a direct timing measurement, not a fitted parameter or self-referential equation. No self-definitional steps, fitted-input predictions, load-bearing self-citations, uniqueness theorems, or ansatz smuggling appear in the provided text. The claims rest on external experimental benchmarks rather than reducing to the paper's own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The framework depends on the generalization ability of existing 3D foundation models to plant data; no new free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption 3D foundation models trained on general scenes can provide accurate feed-forward geometric recovery for plant scenes with self-occlusion and limited view overlap
    Invoked to replace COLMAP-style sparse initialization in the framework description.

pith-pipeline@v0.9.1-grok · 5836 in / 1201 out tokens · 33848 ms · 2026-07-03T16:32:26.010768+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

60 extracted references · 60 canonical work pages · 2 internal anchors

  1. [1]

    IEEE Transactions on Visualization and Computer Graphics , year =

    Bernardini, Fausto and Mittleman, Joshua and Rushmeier, Holly and Silva, Claudio and Taubin, Gabriel , title =. IEEE Transactions on Visualization and Computer Graphics , year =

    work page

  2. [2]

    Proceedings of the European Conference on Computer Vision (ECCV) , year =

    Chen, Liang-Chieh and Zhu, Yukun and Papandreou, George and Schroff, Florian and Adam, Hartwig , title =. Proceedings of the European Conference on Computer Vision (ECCV) , year =

    work page

  3. [3]

    Proceedings of the European Conference on Computer Vision (ECCV) , year =

    Dai, Angela and Niessner, Matthias , title =. Proceedings of the European Conference on Computer Vision (ECCV) , year =

    work page

  4. [4]

    and Bolles, Robert C

    Fischler, Martin A. and Bolles, Robert C. , title =. Communications of the ACM , year =

    work page

  5. [5]

    and Tester, Mark , title =

    Furbank, Robert T. and Tester, Mark , title =. Trends in Plant Science , year =

    work page

  6. [6]

    IEEE Transactions on Pattern Analysis and Machine Intelligence , year =

    Furukawa, Yasutaka and Ponce, Jean , title =. IEEE Transactions on Pattern Analysis and Machine Intelligence , year =

    work page

  7. [7]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Guedon, Antoine and Lepetit, Vincent , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  8. [8]

    Plant Methods , year =

    Harandi, Negin and Vandenberghe, Breght and Vankerschaver, Joris and Depuydt, Stephen and Van Messem, Arnout , title =. Plant Methods , year =

    work page

  9. [9]

    and Holynski, Aleksander and Kanazawa, Angjoo , title =

    Haque, Ayaan and Tancik, Matthew and Efros, Alexei A. and Holynski, Aleksander and Kanazawa, Angjoo , title =. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , year =

    work page

  10. [10]

    and Omholt, Stig , title =

    Houle, David and Govindaraju, Diddahally R. and Omholt, Stig , title =. Nature Reviews Genetics , year =

    work page

  11. [11]

    ACM SIGGRAPH 2024 Conference Papers , year =

    Huang, Binbin and Yu, Zehao and Chen, Anpei and Geiger, Andreas and Gao, Shenghua , title =. ACM SIGGRAPH 2024 Conference Papers , year =

    work page 2024

  12. [12]

    Journal of Robotics and Mechatronics , year =

    Imabuchi, Ryota and Kawabata, Koji , title =. Journal of Robotics and Mechatronics , year =

    work page

  13. [13]

    Remote Sensing , year =

    Itakura, Kenta and Hosoi, Fumiki , title =. Remote Sensing , year =

    work page

  14. [14]

    Frontiers in Plant Science , year =

    Jin, Shichao and Su, Yi and Gao, Shengliang and Wu, Fang and Hu, Tianyu and Liu, Jiayuan and Li, Wei and Wang, Dan and Chen, Shuo and Jiang, Yanli and Pang, Shijie and Guo, Qinghua , title =. Frontiers in Plant Science , year =

    work page

  15. [15]

    ACM Transactions on Graphics , year =

    Kerbl, Bernhard and Kopanas, Georgios and Leimkuhler, Thomas and Drettakis, George , title =. ACM Transactions on Graphics , year =

    work page

  16. [16]

    and Lo, Wan-Yen and Doll

    Kirillov, Alexander and Mintun, Eric and Ravi, Nikhila and Mao, Hanzi and Rolland, Chloe and Gustafson, Laura and Xiao, Tete and Whitehead, Spencer and Berg, Alexander C. and Lo, Wan-Yen and Doll. Segment Anything , booktitle =. 2023 , pages =

    work page 2023

  17. [17]

    Computer Vision -- ECCV 2024 , year =

    Leroy, Vincent and Cabon, Yohann and Revaud, Jerome , title =. Computer Vision -- ECCV 2024 , year =

    work page 2024

  18. [18]

    Plant Phenomics , year =

    Li, Jiajia and Qi, Xinda and Fu, Zongzhe and Wang, Yukun and Xiong, Ge and Zeng, Dong and Wang, Xiaofei and Liu, Xia and Teng, Shuxin and Hosoi, Fumiki and others , title =. Plant Phenomics , year =

    work page

  19. [19]

    Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , year =

    Liu, Ruoshi and Wu, Rundi and Van Hoorick, Basile and Tokmakov, Pavel and Zakharov, Sergey and Vondrick, Carl , title =. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , year =

    work page

  20. [20]

    Computer Vision -- ECCV 2022 , year =

    Long, Xiaoxiao and Lin, Cheng and Wang, Peng and Komura, Taku and Wang, Wenping , title =. Computer Vision -- ECCV 2022 , year =

    work page 2022

  21. [21]

    and Tancik, Matthew and Barron, Jonathan T

    Mildenhall, Ben and Srinivasan, Pratul P. and Tancik, Matthew and Barron, Jonathan T. and Ramamoorthi, Ravi and Ng, Ren , title =. Computer Vision -- ECCV 2020 , year =

    work page 2020

  22. [22]

    and Mildenhall, Ben and Sajjadi, Mehdi S

    Niemeyer, Michael and Barron, Jonathan T. and Mildenhall, Ben and Sajjadi, Mehdi S. M. and Geiger, Andreas and Radwan, Noha , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  23. [23]

    ACM SIGGRAPH Asia 2024 Technical Communications , year =

    Ojo, Tommy and La, Thai and Morton, Andrew and Stavness, Ian , title =. ACM SIGGRAPH Asia 2024 Technical Communications , year =

    work page 2024

  24. [24]

    Breeding Science , year =

    Okura, Fumio , title =. Breeding Science , year =

    work page

  25. [25]

    Medical Image Computing and Computer-Assisted Intervention (MICCAI) , year =

    Ronneberger, Olaf and Fischer, Philipp and Brox, Thomas , title =. Medical Image Computing and Computer-Assisted Intervention (MICCAI) , year =

    work page

  26. [26]

    and Frahm, Jan-Michael , title =

    Schonberger, Johannes L. and Frahm, Jan-Michael , title =. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  27. [27]

    PLoS ONE , year =

    Schunck, David and Magistri, Federico and Rosu, Radu Alexandru and Cornelissen, Jonas and Chebrolu, Nived and Paulus, Stefan and Lottes, Philipp and Stachniss, Cyrill , title =. PLoS ONE , year =

    work page

  28. [28]

    Artificial Intelligence in Agriculture , year =

    Song, Hongli and Wen, Weiliang and Guo, Xinyu , title =. Artificial Intelligence in Agriculture , year =

    work page

  29. [29]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Wang, Shuzhe and Leroy, Vincent and Cabon, Yohann and Chidlovskii, Boris and Revaud, Jerome , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  30. [30]

    $\pi^3$: Permutation-Equivariant Visual Geometry Learning

    Wang, Yifan and Zhou, Jianjun and Zhu, Haoyi and Chang, Wenzheng and Zhou, Yang and Li, Zizun and Chen, Junyi and Pang, Jiangmiao and Shen, Chunhua and He, Tong , title =. arXiv preprint arXiv:2507.13347 , year =

    work page internal anchor Pith review Pith/arXiv arXiv

  31. [31]

    Sensors , year =

    Wang, Yunlong and Zhang, Zhiyong , title =. Sensors , year =

    work page

  32. [32]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Wang, Jianyuan and Chen, Minghao and Karaev, Nikita and Vedaldi, Andrea and Rupprecht, Christian and Novotny, David , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  33. [33]

    Frontiers in Plant Science , year =

    Wu, Sheng and Wen, Weiliang and Xiao, Boxiang and Guo, Xinyu and Du, Jianjun and Wang, Chuanyu and Wang, Yongjian , title =. Frontiers in Plant Science , year =

    work page

  34. [34]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Yang, Jiawei and Pavone, Marco and Wang, Yue , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  35. [35]

    Yu, Zehao and Chen, Anpei and Tancik, Matthew and Charatan, David and Geiger, Andreas and Ng, Ren and Barron, Jonathan T. and Mildenhall, Ben and Hedman, Peter and Kopanas, Georgios and Leimkuhler, Thomas and Drettakis, George and Fan, Lue and Snavely, Noah and Riegler, Gernot and Liu, Songyou and Wang, Qianqian and Dai, Angela , title =. Proceedings of t...

    work page

  36. [36]

    and Lyons, Eric and Pauli, Duke and Benes, Bedrich and Barnard, Kobus , title =

    Zarei, Ariyan and Li, Bosheng and Schnable, James C. and Lyons, Eric and Pauli, Duke and Benes, Bedrich and Barnard, Kobus , title =. Computers and Electronics in Agriculture , year =

    work page

  37. [37]

    International Journal of Remote Sensing , year =

    Zhang, Zhiyong and Zhang, Yong and Ma, Yue and Yang, Chao and Wang, Xiaolong and others , title =. International Journal of Remote Sensing , year =

    work page

  38. [38]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Zhang, Shangzhan and Wang, Jianyuan and Xu, Yinghao and Xue, Nan and Rupprecht, Christian and Zhou, Xiaowei and Shen, Yujun and Wetzstein, Gordon , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =. 2025 , pages =

    work page 2025

  39. [39]

    arXiv preprint arXiv:2511.20343 , year =

    Wang, Hengyi and others , title =. arXiv preprint arXiv:2511.20343 , year =

    work page arXiv

  40. [40]

    Ngo, T. D. and others , title =. arXiv preprint arXiv:2603.03744 , year =

    work page arXiv

  41. [41]

    Chen, L. Z. and others , title =. arXiv preprint arXiv:2604.14141 , year =

    work page internal anchor Pith review Pith/arXiv arXiv

  42. [42]

    Artificial Intelligence in Agriculture , year =

    Yang, Xiaoyue and others , title =. Artificial Intelligence in Agriculture , year =

    work page

  43. [43]

    Plant Phenomics , year =

    Song, Wentao and Huang, He and Sun, Youqiang and Qu, Fang and Zhang, Jiaqi and Fang, Longhui and Hao, Yuwei and Peng, Chenyang , title =. Plant Phenomics , year =

    work page

  44. [44]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops , month =

    Zhang, Daiwei and Gajardo, Joaquin and Medic, Tomislav and Katircioglu, Isinsu and Boss, Mike and Kirchgessner, Norbert and Walter, Achim and Roth, Lukas , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops , month =. 2025 , pages =

    work page 2025

  45. [45]

    Plant Phenomics , year =

    Luo, Liyi and Jiang, Xintong and Yang, Yu and Samy, Eugene Roy Antony and Lefsrud, Mark and Hoyos-Villegas, Valerio and Sun, Shangpeng , title =. Plant Phenomics , year =

    work page

  46. [46]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Wu, Jay Zhangjie and Zhang, Yuxuan and Turki, Haithem and Ren, Xuanchi and Gao, Jun and Shou, Mike Zheng and Fidler, Sanja and Gojcic, Zan and Ling, Huan , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    work page

  47. [47]

    and others , title =

    Hu, Tianyu and Shen, Peirong and Zhang, Y. and others , title =. Plant Methods , year =

    work page

  48. [48]

    The Crop Journal , year =

    Shen, Peng and Xueyao, Jing and Deng, Wenzhe and Jia, Hanyue and Wu, Tingting , title =. The Crop Journal , year =

    work page

  49. [49]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Probing the 3d awareness of visual foundation models , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  50. [50]

    ISPRS Journal of Photogrammetry and Remote Sensing , year =

    Yang, Xin and Miao, Teng and Tian, Xueying and Wang, Dabao and Zhao, Jianxiang and Lin, Lili and Zhu, Chao and Yang, Tao and Xu, Tongyu , title =. ISPRS Journal of Photogrammetry and Remote Sensing , year =

    work page

  51. [51]

    Agriculture , year =

    Sun, Yongzhe and Zhang, Zhixin and Sun, Kai and Li, Shuai and Yu, Jianglin and Miao, Linxiao and Zhang, Zhanguo and Li, Yang and Zhao, Hongjie and Hu, Zhenbang and Xin, Dawei and Chen, Qingyu and Zhu, Rongsheng , title =. Agriculture , year =

    work page

  52. [52]

    and Li, Yawei and Krishnamurthy, Adarsh and Schnable, Patrick S

    Kimara, Elvis and Hadadi, Mozhgan and Godbersen, Jackson and Balu, Aditya and Jubery, Talukder Z. and Li, Yawei and Krishnamurthy, Adarsh and Schnable, Patrick S. and Ganapathysubramanian, Baskar , title =. Plant Phenomics , year =

    work page

  53. [53]

    Plant Phenomics , year =

    Deng, Wenzhe and Wu, Tingting and Ni, Zhichao and Liu, Yajie and Jia, Hanyue and Ling, Qingping , title =. Plant Phenomics , year =

    work page

  54. [54]

    arXiv preprint arXiv:2601.12366 , year =

    Zhang, Jiafei and Cao, Songliang and Xu, Binghui and Li, Yanan and Jia, Weiwei and Wu, Tingting and Lu, Hao and Hu, Weijuan and Han, Zhiguo , title =. arXiv preprint arXiv:2601.12366 , year =

    work page arXiv

  55. [55]

    Plant Phenomics , year =

    Ma, Zhihong and Du, Ruiming and Xie, Jiayang and Sun, Dawei and Fang, Hui and Jiang, Lixi and Cen, Haiyan , title =. Plant Phenomics , year =

    work page

  56. [56]

    aBIOTECH , year =

    Zheng, Yuhui and Gao, Lu and Zhang, Jiafei and Miao, Liming and Zhu, Hongfang and Yang, Xuedong and Zhang, Dingyu and Han, Zhiguo and Li, Xiaofeng and Zhu, Weimin , title =. aBIOTECH , year =

    work page

  57. [57]

    Plant Communications , year =

    Tao, Haiyu and Xu, Sisi and Miao, Chunyu and Long, Haiping and Jin, Shichao and Guo, Qinghua , title =. Plant Communications , year =

    work page

  58. [58]

    and others , title =

    Yang, N. and others , title =. Plant Communications , year =

    work page

  59. [59]

    and Batchelor, William David and Xiong, Lizhong and Yan, Jianbing , title =

    Yang, Wanneng and Feng, Hui and Zhang, Xuehai and Zhang, Jian and Doonan, John H. and Batchelor, William David and Xiong, Lizhong and Yan, Jianbing , title =. Molecular Plant , year =

    work page

  60. [60]

    and Varshney, Rajeev K

    Xu, Yunbi and Zhang, Xingping and Li, Huihui and Zheng, Hong and Zhang, Jian and Olsen, Matthew S. and Varshney, Rajeev K. and Prasanna, Boddupalli M. and Qian, Qian , title =. Molecular Plant , year =

    work page