PiGPDS: Perspective Independent Ground Plane Detection and Segmentation for Complex 3D Indoor Scenes
Bibliographic record
Abstract
Ground plane detection and segmentation techniques can benefit and help improve the accuracy and robustness of a wide range of computer vision applications, from 3D object segmentation and autonomous navigation to mixed and augmented reality. Existing approaches often rely on restrictive assumptions to simplify the problem, such as the ground plane being the largest plane in the scene or the camera location or orientation being ideal. We present a ground plane segmentation technique for real-world 3D indoor scenes where the position and orientation of the sensor are unrestricted and unknown. Our method only requires one 3D point cloud of an indoor scene and assumes that the scene contains at least one surface parallel to the actual ground plane, which is generally true for 3D indoor scenes. We begin by utilizing a voxelized grid downsampling method to enhance the speed of the algorithm. Subsequently, we use K-medoids clustering and an angular-based zone determination technique to identify the ground zone. Next, we divide the ground zone into several clusters using the Euclidean clustering algorithm, and we employ the M-estimator Sample Consensus (MSAC) algorithm to fit the largest plane in each cluster with a specific orientation. Finally, based on the geometric relationship between the fitted planes of the ground zone, we estimate the ground plane, verify it and segment all its associated points using a distance-based approach. We evaluated our method on public and self-generated datasets, in which we positioned a depth sensor at various locations, pitches, and yaws. Our experimental results demonstrate that our proposed method can robustly and efficiently detect and segment the ground plane of complex 3D indoor scenes and supports varied sensor locations and orientations. We evaluate the performance of our proposed method in terms of four conventional metrics: specificity, precision, recall, and F1 score, with average experimental results of 98.28, 95.48, 96.64, and 96.01, respectively.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".