Mohan M. Trivedi

Ross Greer, Nachiket Deo, Akshay Rangesh et al.

To make safe transitions from autonomous to manual control, a vehicle must have a representation of the awareness of driver state; two metrics which quantify this state are the Observable Readiness Index and Takeover Time. In this work, we show that machine learning models which predict these two metrics are robust to multiple camera views, expanding from the limited view angles in prior research. Importantly, these models take as input feature vectors corresponding to hand location and activity as well as gaze location, and we explore the tradeoffs of different views in generating these feature vectors. Further, we introduce two metrics to evaluate the quality of control transitions following the takeover event (the maximal lateral deviation and velocity deviation) and compute correlations of these post-takeover metrics to the pre-takeover predictive metrics.

Ross Greer, Akshay Gopalkrishnan, Nachiket Deo et al.

Detecting road traffic signs and accurately determining how they can affect the driver's future actions is a critical task for safe autonomous driving systems. However, various traffic signs in a driving scene have an unequal impact on the driver's decisions, making detecting the salient traffic signs a more important task. Our research addresses this issue, constructing a traffic sign detection model which emphasizes performance on salient signs, or signs that influence the decisions of a driver. We define a traffic sign salience property and use it to construct the LAVA Salient Signs Dataset, the first traffic sign dataset that includes an annotated salience property. Next, we use a custom salience loss function, Salience-Sensitive Focal Loss, to train a Deformable DETR object detection model in order to emphasize stronger performance on salient signs. Results show that a model trained with Salience-Sensitive Focal Loss outperforms a model trained without, with regards to recall of both salient signs and all signs combined. Further, the performance margin on salient signs compared to all signs is largest for the model trained with Salience-Sensitive Focal Loss.

Ross Greer, Mohan Trivedi

Multi-sensor frameworks provide opportunities for ensemble learning and sensor fusion to make use of redundancy and supplemental information, helpful in real-world safety applications such as continuous driver state monitoring which necessitate predictions even in cases where information may be intermittently missing. We define this problem of intermittent instances of missing information (by occlusion, noise, or sensor failure) and design a learning framework around these data gaps, proposing and analyzing an imputation scheme to handle missing information. We apply these ideas to tasks in camera-based hand activity classification for robust safety during autonomous driving. We show that a late-fusion approach between parallel convolutional neural networks can outperform even the best-placed single camera model in estimating the hands' held objects and positions when validated on within-group subjects, and that our multi-camera framework performs best on average in cross-group validation, and that the fusion approach outperforms ensemble weighted majority and model combination schemes.

Ross Greer, Lulua Rakla, Anish Gopalan et al.

Manual (hand-related) activity is a significant source of crash risk while driving. Accordingly, analysis of hand position and hand activity occupation is a useful component to understanding a driver's readiness to take control of a vehicle. Visual sensing through cameras provides a passive means of observing the hands, but its effectiveness varies depending on camera location. We introduce an algorithmic framework, SMART Hands, for accurate hand classification with an ensemble of camera views using machine learning. We illustrate the effectiveness of this framework in a 4-camera setup, reaching 98% classification accuracy on a variety of locations and held objects for both of the driver's hands. We conclude that this multi-camera framework can be extended to additional tasks such as gaze and pose analysis, with further applications in driver and passenger safety.

Ross Greer, Mohan Trivedi

In this work, we contribute an EM algorithm for estimation of corner points and linear crossing segments for both marked and unmarked pedestrian crosswalks using the detections of pedestrians from processed LiDAR point clouds or camera images. We demonstrate the algorithmic performance by analyzing three real-world datasets containing multiple periods of data collection for four-corner and two-corner intersections with marked and unmarked crosswalks. Additionally, we include a Python video tool to visualize the crossing parameter estimation, pedestrian trajectories, and phase intervals in our public source code.

Ross Greer, Lulua Rakla, Samveed Desai et al.

Vehicles are constantly approaching and sharing the road with pedestrians, and as a result it is critical for vehicles to prevent any collisions with pedestrians. Current methods for pedestrian collision prevention focus on integrating visual pedestrian detectors with Automatic Emergency Braking (AEB) systems which can trigger warnings and apply brakes as a pedestrian enters a vehicle's path. Unfortunately, pedestrian-detection-based systems can be hindered in certain situations such as nighttime or when pedestrians are occluded. Our system, CHAMP (Crowdsourced, History-based Advisories of Mapped Pedestrians), addresses such issues using an online, map-based pedestrian detection system where pedestrian locations are aggregated into a dataset after repeated passes of locations. Using this dataset, we are able to learn pedestrian zones and generate advisory notices when a vehicle is approaching a pedestrian despite challenges like dark lighting or pedestrian occlusion. We collected and carefully annotated pedestrian data in La Jolla, CA to construct training and test sets of pedestrian locations. Moreover, we use the number of correct advisories, false advisories, and missed advisories to define precision and recall performance metrics to evaluate CHAMP. This approach can be tuned such that we achieve a maximum of 100% precision and 75% recall on the experimental dataset, with performance enhancement options through further data collection.

Ross Greer, Akshay Gopalkrishnan, Maitrayee Keskar et al.

This paper explores the representation of vehicle lights in computer vision and its implications for various tasks in the field of autonomous driving. Different specifications for representing vehicle lights, including bounding boxes, center points, corner points, and segmentation masks, are discussed in terms of their strengths and weaknesses. Three important tasks in autonomous driving that can benefit from vehicle light detection are identified: nighttime vehicle detection, 3D vehicle orientation estimation, and dynamic trajectory cues. Each task may require a different representation of the light. The challenges of collecting and annotating large datasets for training data-driven models are also addressed, leading to introduction of the LISA Vehicle Lights Dataset and associated Light Visibility Model, which provides light annotations specifically designed for downstream applications in vehicle detection, intent and trajectory prediction, and safe path planning. A comparison of existing vehicle light datasets is provided, highlighting the unique features and limitations of each dataset. Overall, this paper provides insights into the representation of vehicle lights and the importance of accurate annotations for training effective detection models in autonomous driving applications. Our dataset and model are made available at https://cvrr.ucsd.edu/vehicle-lights-dataset

Akshay Gopalkrishnan, Ross Greer, Maitrayee Keskar et al.

Vehicle light detection, association, and localization are required for important downstream safe autonomous driving tasks, such as predicting a vehicle's light state to determine if the vehicle is making a lane change or turning. Currently, many vehicle light detectors use single-stage detectors which predict bounding boxes to identify a vehicle light, in a manner decoupled from vehicle instances. In this paper, we present a method for detecting a vehicle light given an upstream vehicle detection and approximation of a visible light's center. Our method predicts four approximate corners associated with each vehicle light. We experiment with CNN architectures, data augmentation, and contextual preprocessing methods designed to reduce surrounding-vehicle confusion. We achieve an average distance error from the ground truth corner of 4.77 pixels, about 16.33% of the size of the vehicle light on average. We train and evaluate our model on the LISA Lights Dataset, allowing us to thoroughly evaluate our vehicle light corner detection model on a large variety of vehicle light shapes and lighting conditions. We propose that this model can be integrated into a pipeline with vehicle detection and vehicle light center detection to make a fully-formed vehicle light detection network, valuable to identifying trajectory-informative signals in driving scenes.

Sondos Mohamed, Walter Zimmer, Ross Greer et al.

Accurately detecting 3D objects from monocular images in dynamic roadside scenarios remains a challenging problem due to varying camera perspectives and unpredictable scene conditions. This paper introduces a two-stage training strategy to address these challenges. Our approach initially trains a model on the large-scale synthetic dataset, RoadSense3D, which offers a diverse range of scenarios for robust feature learning. Subsequently, we fine-tune the model on a combination of real-world datasets to enhance its adaptability to practical conditions. Experimental results of the Cube R-CNN model on challenging public benchmarks show a remarkable improvement in detection performance, with a mean average precision rising from 0.26 to 12.76 on the TUM Traffic A9 Highway dataset and from 2.09 to 6.60 on the DAIR-V2X-I dataset when performing transfer learning. Code, data, and qualitative video results are available on the project website: https://roadsense3d.github.io.

Akshay Gopalkrishnan, Ross Greer, Mohan Trivedi

Vision-Language Models (VLMs) and Multi-Modal Language models (MMLMs) have become prominent in autonomous driving research, as these models can provide interpretable textual reasoning and responses for end-to-end autonomous driving safety tasks using traffic scene images and other data modalities. However, current approaches to these systems use expensive large language model (LLM) backbones and image encoders, making such systems unsuitable for real-time autonomous driving systems where tight memory constraints exist and fast inference time is necessary. To address these previous issues, we develop EM-VLM4AD, an efficient, lightweight, multi-frame vision language model which performs Visual Question Answering for autonomous driving. In comparison to previous approaches, EM-VLM4AD requires at least 10 times less memory and floating point operations, while also achieving higher CIDEr and ROUGE-L scores than the existing baseline on the DriveLM dataset. EM-VLM4AD also exhibits the ability to extract relevant information from traffic views related to prompts and can answer questions for various autonomous driving subtasks. We release our code to train and evaluate our model at https://github.com/akshaygopalkr/EM-VLM4AD.

Ross Greer, Akshay Gopalkrishnan, Sumega Mandadi et al.

About 30% of all traffic crash fatalities in the United States involve drunk drivers, making the prevention of drunk driving paramount to vehicle safety in the US and other locations which have a high prevalence of driving while under the influence of alcohol. Driving impairment can be monitored through active use of sensors (when drivers are asked to engage in providing breath samples to a vehicle instrument or when pulled over by a police officer), but a more passive and robust mechanism of sensing may allow for wider adoption and benefit of intelligent systems that reduce drunk driving accidents. This could assist in identifying impaired drivers before they drive, or early in the driving process (before a crash or detection by law enforcement). In this research, we introduce a study which adopts a multi-modal ensemble of visual, thermal, audio, and chemical sensors to (1) examine the impact of acute alcohol administration on driving performance in a driving simulator, and (2) identify data-driven methods for detecting driving under the influence of alcohol. We describe computer vision and machine learning models for analyzing the driver's face in thermal imagery, and introduce a pipeline for training models on data collected from drivers with a range of breath-alcohol content levels, including discussion of relevant machine learning phenomena which can help in future experiment design for related studies.

Ross Greer, Bjørk Antoniussen, Andreas Møgelmose et al.

Object detection is crucial for ensuring safe autonomous driving. However, data-driven approaches face challenges when encountering minority or novel objects in the 3D driving scene. In this paper, we propose VisLED, a language-driven active learning framework for diverse open-set 3D Object Detection. Our method leverages active learning techniques to query diverse and informative data samples from an unlabeled pool, enhancing the model's ability to detect underrepresented or novel objects. Specifically, we introduce the Vision-Language Embedding Diversity Querying (VisLED-Querying) algorithm, which operates in both open-world exploring and closed-world mining settings. In open-world exploring, VisLED-Querying selects data points most novel relative to existing data, while in closed-world mining, it mines novel instances of known classes. We evaluate our approach on the nuScenes dataset and demonstrate its efficiency compared to random sampling and entropy-querying methods. Our results show that VisLED-Querying consistently outperforms random sampling and offers competitive performance compared to entropy-querying despite the latter's model-optimality, highlighting the potential of VisLED for improving object detection in autonomous driving scenarios. We make our code publicly available at https://github.com/Bjork-crypto/VisLED-Querying

Ross Greer, Samveed Desai, Lulua Rakla et al.

It is critical for vehicles to prevent any collisions with pedestrians. Current methods for pedestrian collision prevention focus on integrating visual pedestrian detectors with Automatic Emergency Braking (AEB) systems which can trigger warnings and apply brakes as a pedestrian enters a vehicle's path. Unfortunately, pedestrian-detection-based systems can be hindered in certain situations such as night-time or when pedestrians are occluded. Our system addresses such issues using an online, map-based pedestrian detection aggregation system where common pedestrian locations are learned after repeated passes of locations. Using a carefully collected and annotated dataset in La Jolla, CA, we demonstrate the system's ability to learn pedestrian zones and generate advisory notices when a vehicle is approaching a pedestrian despite challenges like dark lighting or pedestrian occlusion. Using the number of correct advisories, false advisories, and missed advisories to define precision and recall performance metrics, we evaluate our system and discuss future positive effects with further data collection. We have made our code available at https://github.com/s7desai/ped-mapping, and a video demonstration of the CHAMP system at https://youtu.be/dxeCrS_Gpkw.

Walter Zimmer, Akshay Rangesh, Mohan Trivedi

In this paper, we focus on obtaining 2D and 3D labels, as well as track IDs for objects on the road with the help of a novel 3D Bounding Box Annotation Toolbox (3D BAT). Our open source, web-based 3D BAT incorporates several smart features to improve usability and efficiency. For instance, this annotation toolbox supports semi-automatic labeling of tracks using interpolation, which is vital for downstream tasks like tracking, motion planning and motion prediction. Moreover, annotations for all camera images are automatically obtained by projecting annotations from 3D space into the image domain. In addition to the raw image and point cloud feeds, a Masterview consisting of the top view (bird's-eye-view), side view and front views is made available to observe objects of interest from different perspectives. Comparisons of our method with other publicly available annotation tools reveal that 3D annotations can be obtained faster and more efficiently by using our toolbox.

Ross Greer, Mohan Trivedi

This research explores the integration of language embeddings for active learning in autonomous driving datasets, with a focus on novelty detection. Novelty arises from unexpected scenarios that autonomous vehicles struggle to navigate, necessitating higher-level reasoning abilities. Our proposed method employs language-based representations to identify novel scenes, emphasizing the dual purpose of safety takeover responses and active learning. The research presents a clustering experiment using Contrastive Language-Image Pretrained (CLIP) embeddings to organize datasets and detect novelties. We find that the proposed algorithm effectively isolates novel scenes from a collection of subsets derived from two real-world driving datasets, one vehicle-mounted and one infrastructure-mounted. From the generated clusters, we further present methods for generating textual explanations of elements which differentiate scenes classified as novel from other scenes in the data pool, presenting qualitative examples from the clustered results. Our results demonstrate the effectiveness of language-driven embeddings in identifying novel elements and generating explanations of data, and we further discuss potential applications in safe takeovers, data curation, and multi-task active learning.

Ahmed Ghita, Bjørk Antoniussen, Walter Zimmer et al.

The curation of large-scale datasets is still costly and requires much time and resources. Data is often manually labeled, and the challenge of creating high-quality datasets remains. In this work, we fill the research gap using active learning for multi-modal 3D object detection. We propose ActiveAnno3D, an active learning framework to select data samples for labeling that are of maximum informativeness for training. We explore various continuous training methods and integrate the most efficient method regarding computational demand and detection performance. Furthermore, we perform extensive experiments and ablation studies with BEVFusion and PV-RCNN on the nuScenes and TUM Traffic Intersection dataset. We show that we can achieve almost the same performance with PV-RCNN and the entropy-based query strategy when using only half of the training data (77.25 mAP compared to 83.50 mAP) of the TUM Traffic Intersection dataset. BEVFusion achieved an mAP of 64.31 when using half of the training data and 75.0 mAP when using the complete nuScenes dataset. We integrate our active learning framework into the proAnno labeling tool to enable AI-assisted data selection and labeling and minimize the labeling costs. Finally, we provide code, weights, and visualization results on our website: https://active3d-framework.github.io/active3d-framework.

Ross Greer, Bjørk Antoniussen, Mathias V. Andersen et al.

Active learning strategies for 3D object detection in autonomous driving datasets may help to address challenges of data imbalance, redundancy, and high-dimensional data. We demonstrate the effectiveness of entropy querying to select informative samples, aiming to reduce annotation costs and improve model performance. We experiment using the BEVFusion model for 3D object detection on the nuScenes dataset, comparing active learning to random sampling and demonstrating that entropy querying outperforms in most cases. The method is particularly effective in reducing the performance gap between majority and minority classes. Class-specific analysis reveals efficient allocation of annotated resources for limited data budgets, emphasizing the importance of selecting diverse and informative data for model training. Our findings suggest that entropy querying is a promising strategy for selecting data that enhances model learning in resource-constrained environments.

Ross Greer, Mohan Trivedi

This study investigates the use of trajectory and dynamic state information for efficient data curation in autonomous driving machine learning tasks. We propose methods for clustering trajectory-states and sampling strategies in an active learning framework, aiming to reduce annotation and data costs while maintaining model performance. Our approach leverages trajectory information to guide data selection, promoting diversity in the training data. We demonstrate the effectiveness of our methods on the trajectory prediction task using the nuScenes dataset, showing consistent performance gains over random sampling across different data pool sizes, and even reaching sub-baseline displacement errors at just 50% of the data cost. Our results suggest that sampling typical data initially helps overcome the ''cold start problem,'' while introducing novelty becomes more beneficial as the training pool size increases. By integrating trajectory-state-informed active learning, we demonstrate that more efficient and robust autonomous driving systems are possible and practical using low-cost data curation strategies.

Mathias Viborg Andersen, Ross Greer, Andreas Møgelmose et al.

Advanced Driver Assistance Systems (ADAS) in intelligent vehicles rely on accurate driver perception within the vehicle cabin, often leveraging a combination of sensing modalities. However, these modalities operate at varying rates, posing challenges for real-time, comprehensive driver state monitoring. This paper addresses the issue of missing data due to sensor frame rate mismatches, introducing a generative model approach to create synthetic yet realistic thermal imagery. We propose using conditional generative adversarial networks (cGANs), specifically comparing the pix2pix and CycleGAN architectures. Experimental results demonstrate that pix2pix outperforms CycleGAN, and utilizing multi-view input styles, especially stacked views, enhances the accuracy of thermal image generation. Moreover, the study evaluates the model's generalizability across different subjects, revealing the importance of individualized training for optimal performance. The findings suggest the potential of generative models in addressing missing frames, advancing driver state monitoring for intelligent vehicles, and underscoring the need for continued research in model generalization and customization.

Ross Greer, Mathias Viborg Andersen, Andreas Møgelmose et al.

Driver activity classification is crucial for ensuring road safety, with applications ranging from driver assistance systems to autonomous vehicle control transitions. In this paper, we present a novel approach leveraging generalizable representations from vision-language models for driver activity classification. Our method employs a Semantic Representation Late Fusion Neural Network (SRLF-Net) to process synchronized video frames from multiple perspectives. Each frame is encoded using a pretrained vision-language encoder, and the resulting embeddings are fused to generate class probability predictions. By leveraging contrastively-learned vision-language representations, our approach achieves robust performance across diverse driver activities. We evaluate our method on the Naturalistic Driving Action Recognition Dataset, demonstrating strong accuracy across many classes. Our results suggest that vision-language representations offer a promising avenue for driver monitoring systems, providing both accuracy and interpretability through natural language descriptors.

Lars Ullrich, Walter Zimmer, Ross Greer et al.

While artificial intelligence (AI) is advancing rapidly and mastering increasingly complex problems with astonishing performance, the safety assurance of such systems is a major concern. Particularly in the context of safety-critical, real-world cyber-physical systems, AI promises to achieve a new level of autonomy but is hampered by a lack of safety assurance. While data-driven control takes up recent developments in AI to improve control systems, control theory in general could be leveraged to improve AI safety. Therefore, this article outlines a new perspective on AI safety based on an interdisciplinary interpretation of the underlying data-generation process and the respective abstraction by AI systems in a system theory-inspired and system analysis-driven manner. In this context, the new perspective, also referred to as data control, aims to stimulate AI engineering to take advantage of existing safety analysis and assurance in an interdisciplinary way to drive the paradigm of data control. Following a top-down approach, a generic foundation for safety analysis and assurance is outlined at an abstract level that can be refined for specific AI systems and applications and is prepared for future innovation.

Walter Zimmer, Ross Greer, Xingcheng Zhou et al.

Road traffic injuries are the leading cause of death for people aged 5-29, resulting in about 1.19 million deaths each year. To reduce these fatalities, it is essential to address human errors like speeding, drunk driving, and distractions. Additionally, faster accident detection and quicker medical response can help save lives. We propose an accident detection framework that combines a rule-based approach with a learning-based one. We introduce a dataset of real-world highway accidents featuring high-speed crash sequences. It includes 294,924 labeled 2D boxes, 93,012 labeled 3D boxes, and track IDs across 48,144 frames captured at 10 Hz using four roadside cameras and LiDAR sensors. The dataset covers ten object classes and is released in the OpenLABEL format. Our experiments and analysis demonstrate the reliability of our method.

Walter Zimmer, Ross Greer, Xingcheng Zhou et al.

Even though a significant amount of work has been done to increase the safety of transportation networks, accidents still occur regularly. They must be understood as an unavoidable and sporadic outcome of traffic networks. We present the TUM Traffic Accident (TUMTraf-A) dataset, a collection of real-world highway accidents. It contains ten sequences of vehicle crashes at high-speed driving with 294,924 labeled 2D and 93,012 labeled 3D boxes and track IDs within 48,144 labeled frames recorded from four roadside cameras and LiDARs at 10 Hz. The dataset contains ten object classes and is provided in the OpenLABEL format. We propose Accid3nD, an accident detection model that combines a rule-based approach with a learning-based one. Experiments and ablation studies on our dataset show the robustness of our proposed method. The dataset, model, and code are available on our project website: https://tum-traffic-dataset.github.io/tumtraf-a.

Walter Zimmer, Ross Greer, Daniel Lehmberg et al.

Even though a significant amount of work has been done to increase the safety of transportation networks, accidents still occur regularly. They must be understood as unavoidable and sporadic outcomes of traffic networks. No public dataset contains 3D annotations of real-world accidents recorded from roadside camera and LiDAR sensors. We present the TUM Traffic Accid3nD (TUMTraf-Accid3nD) dataset, a collection of real-world highway accidents in different weather and lighting conditions. It contains vehicle crashes at high-speed driving with 2,634,233 labeled 2D bounding boxes, instance masks, and 3D bounding boxes with track IDs. In total, the dataset contains 111,945 labeled image and point cloud frames recorded from four roadside cameras and LiDARs at 25 Hz. The dataset contains six object classes and is provided in the OpenLABEL format. We propose an accident detection model that combines a rule-based approach with a learning-based one. Experiments and ablation studies on our dataset show the robustness of our proposed method. The dataset, model, and code are available on our website: https://accident-dataset.github.io.

Ross Greer, Akshay Gopalkrishnan, Jacob Landgren et al.

One of the most important tasks for ensuring safe autonomous driving systems is accurately detecting road traffic lights and accurately determining how they impact the driver's actions. In various real-world driving situations, a scene may have numerous traffic lights with varying levels of relevance to the driver, and thus, distinguishing and detecting the lights that are relevant to the driver and influence the driver's actions is a critical safety task. This paper proposes a traffic light detection model which focuses on this task by first defining salient lights as the lights that affect the driver's future decisions. We then use this salience property to construct the LAVA Salient Lights Dataset, the first US traffic light dataset with an annotated salience property. Subsequently, we train a Deformable DETR object detection transformer model using Salience-Sensitive Focal Loss to emphasize stronger performance on salient traffic lights, showing that a model trained with this loss function has stronger recall than one trained without.

Ross Greer, Jason Isa, Nachiket Deo et al.

Safe path planning in autonomous driving is a complex task due to the interplay of static scene elements and uncertain surrounding agents. While all static scene elements are a source of information, there is asymmetric importance to the information available to the ego vehicle. We present a dataset with a novel feature, sign salience, defined to indicate whether a sign is distinctly informative to the goals of the ego vehicle with regards to traffic regulations. Using convolutional networks on cropped signs, in tandem with experimental augmentation by road type, image coordinates, and planned maneuver, we predict the sign salience property with 76% accuracy, finding the best improvement using information on vehicle maneuver with sign images.

Akshay Rangesh, Nachiket Deo, Ross Greer et al.

Understanding occupant-vehicle interactions by modeling control transitions is important to ensure safe approaches to passenger vehicle automation. Models which contain contextual, semantically meaningful representations of driver states can be used to determine the appropriate timing and conditions for transfer of control between driver and vehicle. However, such models rely on real-world control take-over data from drivers engaged in distracting activities, which is costly to collect. Here, we introduce a scheme for data augmentation for such a dataset. Using the augmented dataset, we develop and train take-over time (TOT) models that operate sequentially on mid and high-level features produced by computer vision algorithms operating on different driver-facing camera views, showing models trained on the augmented dataset to outperform the initial dataset. The demonstrated model features encode different aspects of the driver state, pertaining to the face, hands, foot and upper body of the driver. We perform ablative experiments on feature combinations as well as model architectures, showing that a TOT model supported by augmented data can be used to produce continuous estimates of take-over times without delay, suitable for complex real-world scenarios.

Akshay Rangesh, Nachiket Deo, Ross Greer et al.

With increasing automation in passenger vehicles, the study of safe and smooth occupant-vehicle interaction and control transitions is key. In this study, we focus on the development of contextual, semantically meaningful representations of the driver state, which can then be used to determine the appropriate timing and conditions for transfer of control between driver and vehicle. To this end, we conduct a large-scale real-world controlled data study where participants are instructed to take-over control from an autonomous agent under different driving conditions while engaged in a variety of distracting activities. These take-over events are captured using multiple driver-facing cameras, which when labelled result in a dataset of control transitions and their corresponding take-over times (TOTs). We then develop and train TOT models that operate sequentially on mid to high-level features produced by computer vision algorithms operating on different driver-facing camera views. The proposed TOT model produces continuous predictions of take-over times without delay, and shows promising qualitative and quantitative results in complex real-world scenarios.

Hala Abualsaud, Sean Liu, David Lu et al.

This study presents an approach to lane detection involving the prediction of binary segmentation masks and per-pixel affinity fields. These affinity fields, along with the binary masks, can then be used to cluster lane pixels horizontally and vertically into corresponding lane instances in a post-processing step. This clustering is achieved through a simple row-by-row decoding process with little overhead; such an approach allows LaneAF to detect a variable number of lanes without assuming a fixed or maximum number of lanes. Moreover, this form of clustering is more interpretable in comparison to previous visual clustering approaches, and can be analyzed to identify and correct sources of error. Qualitative and quantitative results obtained on popular lane detection datasets demonstrate the model's ability to detect and cluster lanes effectively and robustly. Our proposed approach sets a new state-of-the-art on the challenging CULane dataset and the recently introduced Unsupervised LLAMAS dataset.

Akshay Rangesh, Pranav Maheshwari, Mez Gebre et al.

This study follows many classical approaches to multi-object tracking (MOT) that model the problem using dynamic graphical data structures, and adapts this formulation to make it amenable to modern neural networks. Our main contributions in this work are the creation of a framework based on dynamic undirected graphs that represent the data association problem over multiple timesteps, and a message passing graph neural network (MPNN) that operates on these graphs to produce the desired likelihood for every association therein. We also provide solutions and propositions for the computational problems that need to be addressed to create a memory-efficient, real-time, online algorithm that can reason over multiple timesteps, correct previous mistakes, update beliefs, and handle missed/false detections. To demonstrate the efficacy of our approach, we only use the 2D box location and object category ID to construct the descriptor for each object instance. Despite this, our model performs on par with state-of-the-art approaches that make use of additional sensors, as well as multiple hand-crafted and/or learned features. This illustrates that given the right problem formulation and model design, raw bounding boxes (and their kinematics) from any off-the-shelf detector are sufficient to achieve competitive tracking results on challenging MOT benchmarks.

Ross Greer, Nachiket Deo, Mohan Trivedi

Predicting a vehicle's trajectory is an essential ability for autonomous vehicles navigating through complex urban traffic scenes. Bird's-eye-view roadmap information provides valuable information for making trajectory predictions, and while state-of-the-art models extract this information via image convolution, auxiliary loss functions can augment patterns inferred from deep learning by further encoding common knowledge of social and legal driving behaviors. Since human driving behavior is inherently multimodal, models which allow for multimodal output tend to outperform single-prediction models on standard metrics. We propose a loss function which enhances such models by enforcing expected driving rules on all predicted modes. Our contribution to trajectory prediction is twofold; we propose a new metric which addresses failure cases of the off-road rate metric by penalizing trajectories that oppose the ascribed heading (flow direction) of a driving lane, and we show this metric to be differentiable and therefore suitable as an auxiliary loss function. We then use this auxiliary loss to extend the the standard multiple trajectory prediction (MTP) and MultiPath models, achieving improved results on the nuScenes prediction benchmark by predicting trajectories which better conform to the lane-following rules of the road.

Kaouther Messaoud, Nachiket Deo, Mohan M. Trivedi et al.

Predicting the trajectories of surrounding agents is an essential ability for autonomous vehicles navigating through complex traffic scenes. The future trajectories of agents can be inferred using two important cues: the locations and past motion of agents, and the static scene structure. Due to the high variability in scene structure and agent configurations, prior work has employed the attention mechanism, applied separately to the scene and agent configuration to learn the most salient parts of both cues. However, the two cues are tightly linked. The agent configuration can inform what part of the scene is most relevant to prediction. The static scene in turn can help determine the relative influence of agents on each other's motion. Moreover, the distribution of future trajectories is multimodal, with modes corresponding to the agent's intent. The agent's intent also informs what part of the scene and agent configuration is relevant to prediction. We thus propose a novel approach applying multi-head attention by considering a joint representation of the static scene and surrounding agents. We use each attention head to generate a distinct future trajectory to address multimodality of future trajectories. Our model achieves state of the art results on the nuScenes prediction benchmark and generates diverse future trajectories compliant with scene structure and agent configuration.

Akshay Rangesh, Bowen Zhang, Mohan M. Trivedi

A driver's gaze is critical for determining their attention, state, situational awareness, and readiness to take over control from partially automated vehicles. Estimating the gaze direction is the most obvious way to gauge a driver's state under ideal conditions when limited to using non-intrusive imaging sensors. Unfortunately, the vehicular environment introduces a variety of challenges that are usually unaccounted for - harsh illumination, nighttime conditions, and reflective eyeglasses. Relying on head pose alone under such conditions can prove to be unreliable and erroneous. In this study, we offer solutions to address these problems encountered in the real world. To solve issues with lighting, we demonstrate that using an infrared camera with suitable equalization and normalization suffices. To handle eyeglasses and their corresponding artifacts, we adopt image-to-image translation using generative adversarial networks to pre-process images prior to gaze estimation. Our proposed Gaze Preserving CycleGAN (GPCycleGAN) is trained to preserve the driver's gaze while removing potential eyeglasses from face images. GPCycleGAN is based on the well-known CycleGAN approach - with the addition of a gaze classifier and a gaze consistency loss for additional supervision. Our approach exhibits improved performance, interpretability, robustness and superior qualitative results on challenging real-world datasets.

Nachiket Deo, Mohan M. Trivedi

We address the problem of forecasting pedestrian and vehicle trajectories in unknown environments, conditioned on their past motion and scene structure. Trajectory forecasting is a challenging problem due to the large variation in scene structure and the multimodal distribution of future trajectories. Unlike prior approaches that directly learn one-to-many mappings from observed context to multiple future trajectories, we propose to condition trajectory forecasts on plans sampled from a grid based policy learned using maximum entropy inverse reinforcement learning (MaxEnt IRL). We reformulate MaxEnt IRL to allow the policy to jointly infer plausible agent goals, and paths to those goals on a coarse 2-D grid defined over the scene. We propose an attention based trajectory generator that generates continuous valued future trajectories conditioned on state sequences sampled from the MaxEnt policy. Quantitative and qualitative evaluation on the publicly available Stanford drone and NuScenes datasets shows that our model generates trajectories that are diverse, representing the multimodal predictive distribution, and precise, conforming to the underlying scene structure over long prediction horizons.

Daniela Ridel, Nachiket Deo, Denis Wolf et al.

Forecasting long-term human motion is a challenging task due to the non-linearity, multi-modality and inherent uncertainty in future trajectories. The underlying scene and past motion of agents can provide useful cues to predict their future motion. However, the heterogeneity of the two inputs poses a challenge for learning a joint representation of the scene and past trajectories. To address this challenge, we propose a model based on grid representations to forecast agent trajectories. We represent the past trajectories of agents using binary 2-D grids, and the underlying scene as a RGB birds-eye view (BEV) image, with an agent-centric frame of reference. We encode the scene and past trajectories using convolutional layers and generate trajectory forecasts using a Convolutional LSTM (ConvLSTM) decoder. Results on the publicly available Stanford Drone Dataset (SDD) show that our model outperforms prior approaches and outputs realistic future trajectories that comply with scene structure and past motion.

Akshay Rangesh, Mohan M. Trivedi

This paper provides a simple solution for reliably solving image classification tasks tied to spatial locations of salient objects in the scene. Unlike conventional image classification approaches that are designed to be invariant to translations of objects in the scene, we focus on tasks where the output classes vary with respect to where an object of interest is situated within an image. To handle this variant of the image classification task, we propose augmenting the standard cross-entropy (classification) loss with a domain dependent Forced Spatial Attention (FSA) loss, which in essence compels the network to attend to specific regions in the image associated with the desired output class. To demonstrate the utility of this loss function, we consider the task of driver foot activity classification - where each activity is strongly correlated with where the driver's foot is in the scene. Training with our proposed loss function results in significantly improved accuracies, better generalization, and robustness against noise, while obviating the need for very large datasets.

Nachiket Deo, Mohan M. Trivedi

We address multi-modal trajectory forecasting of agents in unknown scenes by formulating it as a planning problem. We present an approach consisting of three models; a goal prediction model to identify potential goals of the agent, an inverse reinforcement learning model to plan optimal paths to each goal, and a trajectory generator to obtain future trajectories along the planned paths. Analysis of predictions on the Stanford drone dataset, shows generalizability of our approach to novel scenes.

Daniela A. Ridel, Nachiket Deo, Denis Wolf et al.

Pedestrians and vehicles often share the road in complex inner city traffic. This leads to interactions between the vehicle and pedestrians, with each affecting the other's motion. In order to create robust methods to reason about pedestrian behavior and to design interfaces of communication between self-driving cars and pedestrians we need to better understand such interactions. In this paper, we present a data-driven approach to implicitly model pedestrians' interactions with vehicles, to better predict pedestrian behavior. We propose a LSTM model that takes as input the past trajectories of the pedestrian and ego-vehicle, and pedestrian head orientation, and predicts the future positions of the pedestrian. Our experiments based on a real-world, inner city dataset captured with vehicle mounted cameras, show that the usage of such cues improve pedestrian prediction when compared to a baseline that purely uses the past trajectory of the pedestrian.

Siddharth, Mohan M. Trivedi

Assessing the driver's attention and detecting various hazardous and non-hazardous events during a drive are critical for driver's safety. Attention monitoring in driving scenarios has mostly been carried out using vision (camera-based) modality by tracking the driver's gaze and facial expressions. It is only recently that bio-sensing modalities such as Electroencephalogram (EEG) are being explored. But, there is another open problem which has not been explored sufficiently yet in this paradigm. This is the detection of specific events, hazardous and non-hazardous, during driving that affects the driver's mental and physiological states. The other challenge in evaluating multi-modal sensory applications is the absence of very large scale EEG data because of the various limitations of using EEG in the real world. In this paper, we use both of the above sensor modalities and compare them against the two tasks of assessing the driver's attention and detecting hazardous vs. non-hazardous driving events. We collect user data on twelve subjects and show how in the absence of very large-scale datasets, we can still use pre-trained deep learning convolution networks to extract meaningful features from both of the above modalities. We used the publicly available KITTI dataset for evaluating our platform and to compare it with previous studies. Finally, we show that the results presented in this paper surpass the previous benchmark set up in the above driver awareness-related applications.

Akshay Rangesh, Mohan M. Trivedi

This paper introduces an approach to produce accurate 3D detection boxes for objects on the ground using single monocular images. We do so by merging 2D visual cues, 3D object dimensions, and ground plane constraints to produce boxes that are robust against small errors and incorrect predictions. First, we train a single-shot convolutional neural network (CNN) that produces multiple visual and geometric cues of interest: 2D bounding boxes, 2D keypoints of interest, coarse object orientations and object dimensions. Subsets of these cues are then used to poll probable ground planes from a pre-computed database of ground planes, to identify the "best fit" plane with highest consensus. Once identified, the "best fit" plane provides enough constraints to successfully construct the desired 3D detection box, without directly predicting the 6DoF pose of the object. The entire ground plane polling (GPP) procedure is constructed as a non-parametrized layer of the CNN that outputs the desired "best fit" plane and the corresponding 3D keypoints, which together define the final 3D bounding box. Doing so allows us to poll thousands of different ground plane configurations without adding considerable overhead, while also creating a single CNN that directly produces the desired output without the need for post processing. We evaluate our method on the 2D detection and orientation estimation benchmark from the challenging KITTI dataset, and provide additional comparisons for 3D metrics of importance. This single-stage, single-pass CNN results in superior localization and orientation estimation compared to more complex and computationally expensive monocular approaches.

Nachiket Deo, Mohan M. Trivedi

Continuous estimation the driver's take-over readiness is critical for safe and timely transfer of control during the failure modes of autonomous vehicles. In this paper, we propose a data-driven approach for estimating the driver's take-over readiness based purely on observable cues from in-vehicle vision sensors. We present an extensive naturalistic drive dataset of drivers in a conditionally autonomous vehicle running on Californian freeways. We collect subjective ratings for the driver's take-over readiness from multiple human observers viewing the sensor feed. Analysis of the ratings in terms of intra-class correlation coefficients (ICCs) shows a high degree of consistency in the ratings across raters. We define a metric for the driver's take-over readiness termed the 'Observable Readiness Index (ORI)' based on the ratings. Finally, we propose an LSTM model for continuous estimation of the driver's ORI based on a holistic representation of the driver's state, capturing gaze, hand, pose and foot activity. Our model estimates the ORI with a mean absolute error of 0.449 on a 5 point scale.

Nachiket Deo, Mohan M. Trivedi

Forecasting the motion of surrounding vehicles is a critical ability for an autonomous vehicle deployed in complex traffic. Motion of all vehicles in a scene is governed by the traffic context, i.e., the motion and relative spatial configuration of neighboring vehicles. In this paper we propose an LSTM encoder-decoder model that uses convolutional social pooling as an improvement to social pooling layers for robustly learning interdependencies in vehicle motion. Additionally, our model outputs a multi-modal predictive distribution over future trajectories based on maneuver classes. We evaluate our model using the publicly available NGSIM US-101 and I-80 datasets. Our results show improvement over the state of the art in terms of RMS values of prediction error and negative log-likelihoods of true future trajectories under the model's predictive distribution. We also present a qualitative analysis of the model's predicted distributions for various traffic scenarios.

Akshay Rangesh, Mohan M. Trivedi

Tasks related to human hands have long been part of the computer vision community. Hands being the primary actuators for humans, convey a lot about activities and intents, in addition to being an alternative form of communication/interaction with other humans and machines. In this study, we focus on training a single feedforward convolutional neural network (CNN) capable of executing many hand related tasks that may be of use in autonomous and semi-autonomous vehicles of the future. The resulting network, which we refer to as HandyNet, is capable of detecting, segmenting and localizing (in 3D) driver hands inside a vehicle cabin. The network is additionally trained to identify handheld objects that the driver may be interacting with. To meet the data requirements to train such a network, we propose a method for cheap annotation based on chroma-keying, thereby bypassing weeks of human effort required to label such data. This process can generate thousands of labeled training samples in an efficient manner, and may be replicated in new environments with relative ease.

Kevan Yuen, Mohan M. Trivedi

In the context of autonomous driving, where humans may need to take over in the event where the computer may issue a takeover request, a key step towards driving safety is the monitoring of the hands to ensure the driver is ready for such a request. This work, focuses on the first step of this process, which is to locate the hands. Such a system must work in real-time and under varying harsh lighting conditions. This paper introduces a fast ConvNet approach, based on the work of original work of OpenPose for full body joint estimation. The network is modified with fewer parameters and retrained using our own day-time naturalistic autonomous driving dataset to estimate joint and affinity heatmaps for driver & passenger's wrist and elbows, for a total of 8 joint classes and part affinity fields between each wrist-elbow pair. The approach runs real-time on real-world data at 40 fps on multiple drivers and passengers. The system is extensively evaluated both quantitatively and qualitatively, showing at least 95% detection performance on joint localization and arm-angle estimation.

Akshay Rangesh, Mohan M. Trivedi

Online multi-object tracking (MOT) is extremely important for high-level spatial reasoning and path planning for autonomous and highly-automated vehicles. In this paper, we present a modular framework for tracking multiple objects (vehicles), capable of accepting object proposals from different sensor modalities (vision and range) and a variable number of sensors, to produce continuous object tracks. This work is a generalization of the MDP framework for MOT, with some key extensions - First, we track objects across multiple cameras and across different sensor modalities. This is done by fusing object proposals across sensors accurately and efficiently. Second, the objects of interest (targets) are tracked directly in the real world. This is a departure from traditional techniques where objects are simply tracked in the image plane. Doing so allows the tracks to be readily used by an autonomous agent for navigation and related tasks. To verify the effectiveness of our approach, we test it on real world highway data collected from a heavily sensorized testbed capable of capturing full-surround information. We demonstrate that our framework is well-suited to track objects through entire maneuvers around the ego-vehicle, some of which take more than a few minutes to complete. We also leverage the modularity of our approach by comparing the effects of including/excluding different sensors, changing the total number of sensors, and the quality of object proposals on the final tracking result.

Siddharth, Akshay Rangesh, Eshed Ohn-Bar et al.

Extracting hand regions and their grasp information from images robustly in real-time is critical for occupants' safety and in-vehicular infotainment applications. It must however, be noted that naturalistic driving scenes suffer from rapidly changing illumination and occlusion. This is aggravated by the fact that hands are highly deformable objects, and change in appearance frequently. This work addresses the task of accurately localizing driver hands and classifying the grasp state of each hand. We use a fast ConvNet to first detect likely hand regions. Next, a pixel-based skin classifier that takes into account the global illumination changes is used to refine the hand detections and remove false positives. This step generates a pixel-level mask for each hand. Finally, we study each such masked regions and detect if the driver is grasping the wheel, or in some cases a mobile phone. Through evaluation we demonstrate that our method can outperform state-of-the-art pixel based hand detectors, while running faster (at 35 fps) than other deep ConvNet based frameworks even for grasp analysis. Hand mask cues are shown to be crucial when analyzing a set of driver hand gestures (wheel/mobile phone grasp and no-grasp) in naturalistic driving settings. The proposed detection and localization pipeline hence can act as a general framework for real-time hand detection and gesture classification.

Sourabh Vora, Akshay Rangesh, Mohan M. Trivedi

Driver gaze has been shown to be an excellent surrogate for driver attention in intelligent vehicles. With the recent surge of highly autonomous vehicles, driver gaze can be useful for determining the handoff time to a human driver. While there has been significant improvement in personalized driver gaze zone estimation systems, a generalized system which is invariant to different subjects, perspectives and scales is still lacking. We take a step towards this generalized system using Convolutional Neural Networks (CNNs). We finetune 4 popular CNN architectures for this task, and provide extensive comparisons of their outputs. We additionally experiment with different input image patches, and also examine how image size affects performance. For training and testing the networks, we collect a large naturalistic driving dataset comprising of 11 long drives, driven by 10 subjects in two different cars. Our best performing model achieves an accuracy of 95.18% during cross-subject testing, outperforming current state of the art techniques for this task. Finally, we evaluate our best performing model on the publicly available Columbia Gaze Dataset comprising of images from 56 subjects with varying head pose and gaze directions. Without any training, our model successfully encodes the different gaze directions on this diverse dataset, demonstrating good generalization capabilities.

Kevan Yuen, Mohan M. Trivedi

A key step to driver safety is to observe the driver's activities with the face being a key step in this process to extracting information such as head pose, blink rate, yawns, talking to passenger which can then help derive higher level information such as distraction, drowsiness, intent, and where they are looking. In the context of driving safety, it is important for the system perform robust estimation under harsh lighting and occlusion but also be able to detect when the occlusion occurs so that information predicted from occluded parts of the face can be taken into account properly. This paper introduces the Occluded Stacked Hourglass, based on the work of original Stacked Hourglass network for body pose joint estimation, which is retrained to process a detected face window and output 68 occlusion heat maps, each corresponding to a facial landmark. Landmark location, occlusion levels and a refined face detection score, to reject false positives, are extracted from these heat maps. Using the facial landmark locations, features such as head pose and eye/mouth openness can be extracted to derive driver attention and activity. The system is evaluated for face detection, head pose, and occlusion estimation on various datasets in the wild, both quantitatively and qualitatively, and shows state-of-the-art results.

Sujitha Martin, Sourabh Vora, Kevan Yuen et al.

The study and modeling of driver's gaze dynamics is important because, if and how the driver is monitoring the driving environment is vital for driver assistance in manual mode, for take-over requests in highly automated mode and for semantic perception of the surround in fully autonomous mode. We developed a machine vision based framework to classify driver's gaze into context rich zones of interest and model driver's gaze behavior by representing gaze dynamics over a time period using gaze accumulation, glance duration and glance frequencies. As a use case, we explore the driver's gaze dynamic patterns during maneuvers executed in freeway driving, namely, left lane change maneuver, right lane change maneuver and lane keeping. It is shown that condensing gaze dynamics into durations and frequencies leads to recurring patterns based on driver activities. Furthermore, modeling these patterns show predictive powers in maneuver detection up to a few hundred milliseconds a priori.

Akshay Rangesh, Mohan M. Trivedi

The intelligent vehicle community has devoted considerable efforts to model driver behavior, and in particular to detect and overcome driver distraction in an effort to reduce accidents caused by driver negligence. However, as the domain increasingly shifts towards autonomous and semi-autonomous solutions, the driver is no longer integral to the decision making process, indicating a need to refocus efforts elsewhere. To this end, we propose to study pedestrian distraction instead. In particular, we focus on detecting pedestrians who are engaged in secondary activities involving their cellphones and similar handheld multimedia devices from a purely vision-based standpoint. To achieve this objective, we propose a pipeline incorporating articulated human pose estimation, followed by a soft object label transfer from an ensemble of exemplar SVMs trained on the nearest neighbors in pose feature space. We additionally incorporate head gaze features and prior pose information to carry out cellphone related pedestrian activity recognition. Finally, we offer a method to reliably track the articulated pose of a pedestrian through a sequence of images using a particle filter with a Gaussian Process Dynamical Model (GPDM), which can then be used to estimate sequentially varying activity scores at a very low computational cost. The entire framework is fast (especially for sequential data) and accurate, and easily extensible to include other secondary activities and sources of distraction.