Regularizing Neural Networks for BEV Semantic Segmentation via Inter-class Hierarchy and Spatially-aware Weight Adjustment
Bibliographic record
Abstract
In this work, we propose an effective multi-label semantic segmentation framework for Bird’s Eye View (BEV) perception. While existing BEV frameworks typically employ a separate model for binary segmentation of each semantic class—achieving state-of-the-art performance per class—this design is impractical for real-world autonomous driving applications. A more feasible solution for deployment demands a single model capable of performing multi-label or multi-class segmentation across multiple object categories. To this end, we introduce two key strategies that enhance segmentation quality without modifying the architecture of existing BEV models. First, we incorporate multi-class prediction to capture interclass hierarchies, allowing the model to learn dependencies between semantic categories such as roads, vehicles, and pedestrians. This improves semantic reasoning and boundary delineation, especially in spatially overlapping regions. Second, we propose a spatially-aware weight adjustment (SAWA) that emphasizes rare object zones on the BEV map. This addresses the inherent class imbalance and spatial sparsity of BEV segmentation tasks.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".