# Is my Driver Observation Model Overconfident? Input-guided Calibration Networks for Reliable and Interpretable Confidence Estimates

arXiv: https://arxiv.org/abs/2204.04674 · HTML: https://mlsanity.com/paper/2204.04674/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 86 references that are in the map.
- First posted on arXiv 10 Apr 2022.

## Builds on (most cited first, 28 of 28)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images](https://mlsanity.com/paper/1412.1897/) (2014)
- [Measuring Calibration in Deep Learning](https://mlsanity.com/paper/1904.01685/) (2019)
- [Learning Spatio-Temporal Representation with Pseudo-3D Residual Networks](https://mlsanity.com/paper/1711.10305/) (2017)
- [Calibrating Deep Neural Networks using Focal Loss](https://mlsanity.com/paper/2002.09437/) (2020)
- [Deep Multi-modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges](https://mlsanity.com/paper/1902.07830/) (2019)
- [Analyzing Uncertainty in Neural Machine Translation](https://mlsanity.com/paper/1803.00047/) (2018)
- [The Limits and Potentials of Deep Learning for Robotics](https://mlsanity.com/paper/1804.06557/) (2018)
- [Explainability of deep vision-based autonomous driving systems: Review and challenges](https://mlsanity.com/paper/2101.05307/) (2021)
- [DADA: Driver Attention Prediction in Driving Accident Scenarios](https://mlsanity.com/paper/1912.12148/) (2019)
- [An Analysis of Adversarial Attacks and Defenses on Autonomous Driving Models](https://mlsanity.com/paper/2002.02175/) (2020)
- [Resource Efficient 3D Convolutional Neural Networks](https://mlsanity.com/paper/1904.02422/) (2019)
- [DMD: A Large-Scale Multi-Modal Driver Monitoring Dataset for Attention and Alertness Analysis](https://mlsanity.com/paper/2008.12085/) (2020)
- [DensePASS: Dense Panoramic Semantic Segmentation via Unsupervised Domain Adaptation with Attention-Augmented Context Exchange](https://mlsanity.com/paper/2108.06383/) (2021)
- [Real-time Distracted Driver Posture Classification](https://mlsanity.com/paper/1706.09498/) (2017)
- [MASS: Multi-Attentional Semantic Segmentation of LiDAR Data for Dense Top-View Understanding](https://mlsanity.com/paper/2107.00346/) (2021)
- [Coarse Temporal Attention Network (CTA-Net) for Driver's Activity Recognition](https://mlsanity.com/paper/2101.06636/) (2021)
- [Looking at the Driver/Rider in Autonomous Vehicles to Predict Take-Over Readiness](https://mlsanity.com/paper/1811.06047/) (2018)
- [DriverMHG: A Multi-Modal Dataset for Dynamic Recognition of Driver Micro Hand Gestures and a Real-Time Recognition Framework](https://mlsanity.com/paper/2003.00951/) (2020)
- [Driver Action Prediction Using Deep (Bidirectional) Recurrent Neural Network](https://mlsanity.com/paper/1706.02257/) (2017)
- [Uncertainty-sensitive Activity Recognition: a Reliability Benchmark and the CARING Models](https://mlsanity.com/paper/2101.00468/) (2021)

## Built on by (most cited first, 4 of 4)

- [Vision-Language Models can Identify Distracted Driver Behavior from Naturalistic Videos](https://mlsanity.com/paper/2306.10159/) (2023)
- [Boldness-Recalibration for Binary Event Predictions](https://mlsanity.com/paper/2305.03780/) (2023)
- [On Transferability of Driver Observation Models from Simulated to Real Environments in Autonomous Cars](https://mlsanity.com/paper/2307.16543/) (2023)
- [Quantized Distillation: Optimizing Driver Activity Recognition Models for Resource-Constrained Environments](https://mlsanity.com/paper/2311.05970/) (2023)

## Abstract (arXiv)

Driver observation models are rarely deployed under perfect conditions. In practice, illumination, camera placement and type differ from the ones present during training and unforeseen behaviours may occur at any time. While observing the human behind the steering wheel leads to more intuitive human-vehicle-interaction and safer driving, it requires recognition algorithms which do not only predict the correct driver state, but also determine their prediction quality through realistic and interpretable confidence measures. Reliable uncertainty estimates are crucial for building trust and are a serious obstacle for deploying activity recognition networks in real driving systems. In this work, we for the first time examine how well the confidence values of modern driver observation models indeed match the probability of the correct outcome and show that raw neural network-based approaches tend to significantly overestimate their prediction quality. To correct this misalignment between the confidence values and the actual uncertainty, we consider two strategies. First, we enhance two activity recognition models often used for driver observation with temperature scaling-an off-the-shelf method for confidence calibration in image classification. Then, we introduce Calibrated Action Recognition with Input Guidance (CARING)-a novel approach leveraging an additional neural network to learn scaling the confidences depending on the video representation. Extensive experiments on the Drive&Act dataset demonstrate that both strategies drastically improve the quality of model confidences, while our CARING model out-performs both, the original architectures and their temperature scaling enhancement, leading to best uncertainty estimates.