Underwater Monocular-continuous Stereo Network Based on Cascade Structure for Underwater Image Depth Estimation
Bibliographic record
Abstract
Underwater monocular image depth estimation (UMIDE) is crucial accurately representing and understanding underwater spatial variations, which can significantly enhance applications such as ocean engineering construction and seabed resource exploration. However, UMIDE frequently suffers from isolated discontinuous irregular "spots", inaccurate or indistinguishable edges, and limited model generalization, resulting from color distortion, image blurring, and spatial information loss. This paper proposes an underwater Monocular-continuous stereo network based on a cascade structure (UMCS-CS). Initially, we design a Pinhole model-based Structure from Motion method for camera pose estimation. UMCS-CS employs a two-stage structure for feature extraction: the first stage extracts global information, and the second stage captures detailed information using the squeeze–excitation block with spatial and channel attention. For isolated, discontinuous, and irregular "spots", we use the variance of the current depth estimation to adjust and appropriately expand the depth estimation range. We design a composite loss function, which is a combination of the smooth L1 loss, edge loss function, structural similarity loss, and smoothness loss functions, each with different weights. Experiments on public underwater datasets show that the relative error of the estimated depth map is reduced by 60.83%, the root mean square error by 54.87%, and the logarithmic error by 39.61%.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 source (direct Gemma or distilled Codex), 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".