Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 24, 2026Smart Agricultural TechnologyOpen Access

Enhancing annotations for 5D apple pose estimation through 3D Gaussian Splatting (3DGS)

View Full Paper
Ask AI
Bookmark
Share

Authors

RVRobert van de VenTBTrim BresillaBNBram Nelissen

Discussion

Loading...

Member takes

Overview

Novel pipeline streamlines annotations and enhances pose estimation in apples, indicating significant improvements.

Key Points

  • The research aims to enhance the efficiency and accuracy of apple pose estimation through improved annotation methods.
  • Developed a novel pipeline utilizing 3D Gaussian Splatting for scene reconstruction.
  • Automated the projection of simplified annotations to images.
  • Quantified occlusion rates during training for optimized label performance.
  • Evaluated pose estimation with various occlusion rates and dataset sizes.
  • Achieved a 99.6% reduction in manual annotations required.
  • Best performance observed with training labels from fruits ≤ 95% occluded.
  • Neutral F1 score of 0.927 on original images and 0.970 on rendered images.
  • Least occluded fruits exhibited the best position estimation accuracy.

Cite This Study

Ven et al. (2026) studied this question.

synapsesocial.com/papers/699d3fe6de8e28729cf64c7fhttps://doi.org/10.1016/j.atech.2026.101910
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Adapting SAM3 for 3D fruit counting with cross-view contrastive learning and Hough voting2026
  2. 2A Novel YOLO-Like Multi-Branch Architecture for Accurate Apple Detection and Segmentation Under Orchard Constraints2025
  3. 3Occluded apples orientation estimator based on deep learning model for robotic harvesting2024 · 37 citations
  4. 4Geo-Referenced Factor-Graph SLAM for Orchard-Scale 3D Apple Reconstruction and Yield Estimation2026
  5. 5Dual-Detector Vision and Depth-Aware Back-Projection for Accurate Apple Detection and 3D Localisation for Robotic Harvesting2026