Occlusion-aware point selection and fusion for understory segmentation
- Label
- P1 · ICCV 2027
- Venue
- ICCV 2027 (fallback: CVPR 2027 EarthVision workshop or WACV 2028)
- Status
- planning
- Abstract
- 1 Mar 2027 (est.)
- Deadline
- 8 Mar 2027 (est.)
- Slack
- 87 days before the 14-day buffer
At an equal compute budget, does a point-selection rule that accounts for occlusion keep more sub-canopy evidence and segment the understory better than standard sampling?
Experiments
Tasks
- 27 Nov 2026 · E1 sampler against random, voxel, FPS, default and height-stratified · Planned
- 30 Nov 2026 · E1 go / no-go against height-stratified selection · Planned (milestone)
- 1 Mar 2027 · Register the ICCV abstract (est.) · Planned (milestone)
- 8 Mar 2027 · Submit the ICCV paper (est.) · Planned (milestone)