Benchmarking Lightweight Deep Learning Models for In-Vehicle Face Anti-Spoofing
Bibliographic record
Abstract
In-vehicle face anti-spoofing (FAS) is a crucial security requirement for automotive facial authentication systems in modern cars. Given the demanding cabin environments and limited computational resources, VFPAD (in-Vehicle Face Presentation Attack Detection) dataset, a benchmark for NIR (near infrared)-based in-vehicle presentation attack detection, is used in this case study to evaluate six lightweight deep learning architectures: FeatherNetB, MobileNetV3-small, MiniFASNetV2, MobileNetV4-small, MobileViTV3-XS, and EfficientNetV2-B0. Using a consistent training protocol with transfer learning and fine-tuning, we assess each model according to its classification performance (average classification error rate (ACER), accuracy, and F1 score) and efficiency (floating point operations per second (FLOPS) and number of model parameters). According to experimental results, more recent mobile-oriented deep neural networks (DNNs)-MobileNetV4 and EfficientNetV2-B0 in particular-perform better than older and transformer-based designs, achieving ACERs as low as 0.0086 while keeping latency and model sizes small. Our results highlight the feasibility of implementing real-time FAS systems in vehicular environments with contemporary lightweight architectures.
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) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".