HiSS: Human-Inspired Semantic Segmentation for Vehicle Interior Scene Understanding
Bibliographic record
Abstract
Computer vision plays a critical role in autonomous vehicle technology, with semantic segmentation being a key component by labeling each pixel in an image. This study evaluates the effectiveness of a bio-inspired hybrid approach that combines binary and multiclass segmentation process to improve semantic segmentation in vehicle interiors as part of a Cabin Monitoring System, drawing inspiration from human visual perception processes. Experiments were conducted on synthetic data from the SVIRO dataset using SegFormer, a transformer-based model for segmentation, and YOLO for detection, respectively. Various architectures with a novel hybrid approach were tested that mimics the hierarchical processing of the human attention mechanism by using different Gaussian and distance transform-based masks to refine the focus of the multiclass model. The innovative hybrid approach combining binary and multiclass segmentation improved the mean Intersection over Union from 0.55 to 0.61. Two perceptually-motivated postprocessing methods were developed: a basic approach inspired by visual completion phenomena that further enhanced mIoU to 0.76, and an extended context-aware variant that achieved better accuracy (0.87 vs. 0.84) by analyzing neighborhood information to determine appropriate replacement classes, simi-lar to human contextual integration processes. These methods are highly generic and can be adapted to a wide range of segmentation challenges beyond vehicle interiors. The integration of human perceptual principles into computational vision systems demonstrates the value of biomimetic approaches in advancing artificial intelligence. In addition, we provide a detailed discussion of the results and outline potential directions for future research.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".