Gaurav Pandey

Zhengxia Zou, Rusheng Zhang, Shengyin Shen et al.

We propose a novel and pragmatic framework for traffic scene perception with roadside cameras. The proposed framework covers a full-stack of roadside perception pipeline for infrastructure-assisted autonomous driving, including object detection, object localization, object tracking, and multi-camera information fusion. Unlike previous vision-based perception frameworks rely upon depth offset or 3D annotation at training, we adopt a modular decoupling design and introduce a landmark-based 3D localization method, where the detection and localization can be well decoupled so that the model can be easily trained based on only 2D annotations. The proposed framework applies to either optical or thermal cameras with pinhole or fish-eye lenses. Our framework is deployed at a two-lane roundabout located at Ellsworth Rd. and State St., Ann Arbor, MI, USA, providing 7x24 real-time traffic flow monitoring and high-precision vehicle trajectory extraction. The whole system runs efficiently on a low-power edge computing device with all-component end-to-end delay of less than 20ms.

Sri Aditya Deevi, Connor Lee, Lu Gan et al.

Multimodal deep sensor fusion has the potential to enable autonomous vehicles to visually understand their surrounding environments in all weather conditions. However, existing deep sensor fusion methods usually employ convoluted architectures with intermingled multimodal features, requiring large coregistered multimodal datasets for training. In this work, we present an efficient and modular RGB-X fusion network that can leverage and fuse pretrained single-modal models via scene-specific fusion modules, thereby enabling joint input-adaptive network architectures to be created using small, coregistered multimodal datasets. Our experiments demonstrate the superiority of our method compared to existing works on RGB-thermal and RGB-gated datasets, performing fusion using only a small amount of additional parameters. Our code is available at https://github.com/dsriaditya999/RGBXFusion.

Ella Rabinovich, Matan Vetzler, David Boaz et al. · ibm-research

The rapidly growing market demand for automatic dialogue agents capable of goal-oriented behavior has caused many tech-industry leaders to invest considerable efforts into task-oriented dialog systems. The success of these systems is highly dependent on the accuracy of their intent identification -- the process of deducing the goal or meaning of the user's request and mapping it to one of the known intents for further processing. Gaining insights into unrecognized utterances -- user requests the systems fail to attribute to a known intent -- is therefore a key process in continuous improvement of goal-oriented dialog systems. We present an end-to-end pipeline for processing unrecognized user utterances, deployed in a real-world, commercial task-oriented dialog system, including a specifically-tailored clustering algorithm, a novel approach to cluster representative extraction, and cluster naming. We evaluated the proposed components, demonstrating their benefits in the analysis of unrecognized user requests.

Gaurav Pandey, Danish Contractor, Sachindra Joshi · ibm-research

Embedding-based approaches for dialog response retrieval embed the context-response pairs as points in the embedding space. These approaches are scalable, but fail to account for the complex, many-to-many relationships that exist between context-response pairs. On the other end of the spectrum, there are approaches that feed the context-response pairs jointly through multiple layers of neural networks. These approaches can model the complex relationships between context-response pairs, but fail to scale when the set of responses is moderately large (>100). In this paper, we combine the best of both worlds by proposing a scalable model that can learn complex relationships between context-response pairs. Specifically, the model maps the contexts as well as responses to probability distributions over the embedding space. We train the models by optimizing the Kullback-Leibler divergence between the distributions induced by context-response pairs in the training data. We show that the resultant model achieves better performance as compared to other embedding-based approaches on publicly available conversation data.

Shenbagaraj Kannapiran, Nalin Bendapudi, Ming-Yuan Yu et al.

Robust feature matching forms the backbone for most Visual Simultaneous Localization and Mapping (vSLAM), visual odometry, 3D reconstruction, and Structure from Motion (SfM) algorithms. However, recovering feature matches from texture-poor scenes is a major challenge and still remains an open area of research. In this paper, we present a Stereo Visual Odometry (StereoVO) technique based on point and line features which uses a novel feature-matching mechanism based on an Attention Graph Neural Network that is designed to perform well even under adverse weather conditions such as fog, haze, rain, and snow, and dynamic lighting conditions such as nighttime illumination and glare scenarios. We perform experiments on multiple real and synthetic datasets to validate the ability of our method to perform StereoVO under low visibility weather and lighting conditions through robust point and line matches. The results demonstrate that our method achieves more line feature matches than state-of-the-art line matching algorithms, which when complemented with point feature matches perform consistently well in adverse weather and dynamic lighting conditions.

Sonam Gupta, Yatin Nandwani, Asaf Yehudai et al. · ibm-research

Fine-tuning Large Language Models (LLMs) on specific datasets is a common practice to improve performance on target tasks. However, this performance gain often leads to overfitting, where the model becomes too specialized in either the task or the characteristics of the training data, resulting in a loss of generalization. This paper introduces Selective Self-Rehearsal (SSR), a fine-tuning approach that achieves performance comparable to the standard supervised fine-tuning (SFT) while improving generalization. SSR leverages the fact that there can be multiple valid responses to a query. By utilizing the model's correct responses, SSR reduces model specialization during the fine-tuning stage. SSR first identifies the correct model responses from the training set by deploying an appropriate LLM as a judge. Then, it fine-tunes the model using the correct model responses and the gold response for the remaining samples. The effectiveness of SSR is demonstrated through experiments on the task of identifying unanswerable queries across various datasets. The results show that standard SFT can lead to an average performance drop of up to $16.7\%$ on multiple benchmarks, such as MMLU and TruthfulQA. In contrast, SSR results in close to $2\%$ drop on average, indicating better generalization capabilities compared to standard SFT.

Pardis Taghavi, Reza Langari, Gaurav Pandey

This research paper presents an innovative multi-task learning framework that allows concurrent depth estimation and semantic segmentation using a single camera. The proposed approach is based on a shared encoder-decoder architecture, which integrates various techniques to improve the accuracy of the depth estimation and semantic segmentation task without compromising computational efficiency. Additionally, the paper incorporates an adversarial training component, employing a Wasserstein GAN framework with a critic network, to refine model's predictions. The framework is thoroughly evaluated on two datasets - the outdoor Cityscapes dataset and the indoor NYU Depth V2 dataset - and it outperforms existing state-of-the-art methods in both segmentation and depth estimation tasks. We also conducted ablation studies to analyze the contributions of different components, including pre-training strategies, the inclusion of critics, the use of logarithmic depth scaling, and advanced image augmentations, to provide a better understanding of the proposed framework. The accompanying source code is accessible at \url{https://github.com/PardisTaghavi/SwinMTL}.

Jamie J. R. Bennett, Aviad Susman, Yan Chak Li et al.

In this paper, we introduce eipy--an open-source Python package for developing effective, multi-modal heterogeneous ensembles for classification. eipy simultaneously provides both a rigorous, and user-friendly framework for comparing and selecting the best-performing multi-modal data integration and predictive modeling methods by systematically evaluating their performance using nested cross-validation. The package is designed to leverage scikit-learn-like estimators as components to build multi-modal predictive models. An up-to-date user guide, including API reference and tutorials, for eipy is maintained at https://eipy.readthedocs.io . The main repository for this project can be found on GitHub at https://github.com/GauravPandeyLab/eipy .

Vivek Anand, Bharat Lohani, Rakesh Mishra et al.

Reliable LiDAR perception requires robustness across sensors, environments, and adverse weather. However, existing datasets rarely provide physically consistent observations of the same scene under varying sensor configurations and weather conditions, limiting systematic analysis of domain shifts. This work presents ReaLiTy, a unified physics-informed framework that transforms LiDAR data to match target sensor specifications and weather conditions. The framework integrates physically grounded cues with a learning-based module to generate realistic intensity patterns, while a physics-based weather model introduces consistent geometric and radiometric degradations. Building on this framework, we introduce the LiDAR Adaptation Dataset Suite (LADS), a collection of physically consistent, transformation-ready point clouds with one-to-one correspondence to original datasets. Experiments demonstrate improved cross-domain consistency and realistic weather effects. ReaLiTy and LADS provide a reproducible foundation for studying LiDAR adaptation and simulation-driven perception in intelligent transportation systems.

Subodh Mishra, Gaurav Pandey, Srikanth Saripalli

This work presents a novel target-free extrinsic calibration algorithm for a 3D Lidar and an IMU pair using an Extended Kalman Filter (EKF) which exploits the \textit{motion based calibration constraint} for state update. The steps include, data collection by motion excitation of the Lidar Inertial Sensor suite along all degrees of freedom, determination of the inter sensor rotation by using rotational component of the aforementioned \textit{motion based calibration constraint} in a least squares optimization framework, and finally, the determination of inter sensor translation using the \textit{motion based calibration constraint} for state update in an Extended Kalman Filter (EKF) framework. We experimentally validate our method using data collected in our lab and open-source (https://github.com/unmannedlab/imu_lidar_calibration) our contribution for the robotics research community.

Ana Stanescu, Gaurav Pandey

Heterogeneous ensembles that can aggregate an unrestricted number and variety of base predictors can effectively address challenging prediction problems. In particular, accurate ensembles that are also parsimonious, i.e., consist of as few base predictors as possible, can help reveal potentially useful knowledge about the target problem domain. Although ensemble selection offers a potential approach to achieving these goals, the currently available algorithms are limited in their abilities. In this paper, we present several algorithms that incorporate ensemble diversity into a reinforcement learning (RL)-based ensemble selection framework to build accurate and parsimonious ensembles. These algorithms, as well as several baselines, are rigorously evaluated on datasets from diverse domains in terms of the predictive performance and parsimony of their ensembles. This evaluation demonstrates that our diversity-incorporated RL-based algorithms perform better than the others for constructing simultaneously accurate and parsimonious ensembles. These algorithms can eventually aid the interpretation or reverse engineering of predictive models assimilated into effective ensembles. To enable such a translation, an implementation of these algorithms, as well the experimental setup they are evaluated in, has been made available at https://github.com/GauravPandeyLab/lens-learning-ensembles-using-reinforcement-learning.

Siddharth Agarwal, Ankit Vora, Gaurav Pandey et al.

This paper presents a challenging multi-agent seasonal dataset collected by a fleet of Ford autonomous vehicles at different days and times during 2017-18. The vehicles traversed an average route of 66 km in Michigan that included a mix of driving scenarios such as the Detroit Airport, freeways, city-centers, university campus and suburban neighbourhoods, etc. Each vehicle used in this data collection is a Ford Fusion outfitted with an Applanix POS-LV GNSS system, four HDL-32E Velodyne 3D-lidar scanners, 6 Point Grey 1.3 MP Cameras arranged on the rooftop for 360-degree coverage and 1 Pointgrey 5 MP camera mounted behind the windshield for the forward field of view. We present the seasonal variation in weather, lighting, construction and traffic conditions experienced in dynamic urban environments. This dataset can help design robust algorithms for autonomous vehicles and multi-agent systems. Each log in the dataset is time-stamped and contains raw data from all the sensors, calibration values, pose trajectory, ground truth pose, and 3D maps. All data is available in Rosbag format that can be visualized, modified and applied using the open-source Robot Operating System (ROS). We also provide the output of state-of-the-art reflectivity-based localization for bench-marking purposes. The dataset can be freely downloaded at our website.

Vivek Anand, Bharat Lohani, Rakesh Mishra et al.

Accurate LiDAR simulation is crucial for autonomous driving, especially under adverse weather conditions. Existing methods struggle to capture the complex interactions between LiDAR signals and atmospheric phenomena, leading to unrealistic representations. This paper presents a physics-informed learning framework (PICWGAN) for generating realistic LiDAR data under adverse weather conditions. By integrating physicsdriven constraints for modeling signal attenuation and geometryconsistent degradations into a physics-informed learning pipeline, the proposed method reduces the sim-to-real gap. Evaluations on real-world datasets (CADC for snow, Boreas for rain) and the VoxelScape dataset show that our approach closely mimics realworld intensity patterns. Quantitative metrics, including MSE, SSIM, KL divergence, and Wasserstein distance, demonstrate statistically consistent intensity distributions. Additionally, models trained on data enhanced by our framework outperform baselines in downstream 3D object detection, achieving performance comparable to models trained on real-world data. These results highlight the effectiveness of the proposed approach in improving the realism of LiDAR data and enabling robust perception under adverse weather conditions.

Gaurav Pandey, Gregory Stevens, Henry Liu

This paper presents the functional safety analysis for an Infrastructure-Enabled Depot Autonomy (IX-DA) system. The IX-DA system automates the marshalling of delivery vehicles within a controlled depot environment, navigating connected autonomous vehicles (CAVs) between drop-off zones, service stations (washing, calibration, charging, loading), and pick-up zones without human intervention. We describe the system architecture comprising three principal subsystems -- the connected autonomous vehicle, the infrastructure sensing and compute layer, and the human operator interface -- and derive their functional requirements. Using ISO 26262-compliant Hazard Analysis and Risk Assessment (HARA) methodology, we identify eight hazardous events, evaluate them across different operating scenarios, and assign Automotive Safety Integrity Levels~(ASILs) ranging from Quality Management (QM) to ASIL C. Six safety goals are derived and allocated to vehicle and infrastructure subsystems. The analysis demonstrates that high-speed uncontrolled operation imposes the most demanding safety requirements (ASIL C), while controlled low-speed operation reduces most goals to QM, offering a practical pathway for phased deployment.

Peng Jiang, Gaurav Pandey, Srikanth Saripalli

This paper presents a novel system designed for 3D mapping and visual relocalization using 3D Gaussian Splatting. Our proposed method uses LiDAR and camera data to create accurate and visually plausible representations of the environment. By leveraging LiDAR data to initiate the training of the 3D Gaussian Splatting map, our system constructs maps that are both detailed and geometrically accurate. To mitigate excessive GPU memory usage and facilitate rapid spatial queries, we employ a combination of a 2D voxel map and a KD-tree. This preparation makes our method well-suited for visual localization tasks, enabling efficient identification of correspondences between the query image and the rendered image from the Gaussian Splatting map via normalized cross-correlation (NCC). Additionally, we refine the camera pose of the query image using feature-based matching and the Perspective-n-Point (PnP) technique. The effectiveness, adaptability, and precision of our system are demonstrated through extensive evaluation on the KITTI360 dataset.

Kushagra Bhushan, Yatin Nandwani, Dinesh Khandelwal et al. · ibm-research

Retrieval-Augmented Generation (RAG) has emerged as a prominent method for incorporating domain knowledge into Large Language Models (LLMs). While RAG enhances response relevance by incorporating retrieved domain knowledge in the context, retrieval errors can still lead to hallucinations and incorrect answers. To recover from retriever failures, domain knowledge is injected by fine-tuning the model to generate the correct response, even in the case of retrieval errors. However, we observe that without systematic knowledge augmentation, fine-tuned LLMs may memorize new information but still fail to extract relevant domain knowledge, leading to poor performance. In this work, we present a novel framework that significantly enhances the fine-tuning process by augmenting the training data in two ways -- context augmentation and knowledge paraphrasing. In context augmentation, we create multiple training samples for a given QA pair by varying the relevance of the retrieved information, teaching the model when to ignore and when to rely on retrieved content. In knowledge paraphrasing, we fine-tune with multiple answers to the same question, enabling LLMs to better internalize specialized knowledge. To mitigate catastrophic forgetting due to fine-tuning, we add a domain-specific identifier to a question and also utilize a replay buffer containing general QA pairs. Experimental results demonstrate the efficacy of our method over existing techniques, achieving up to 10\% relative gain in token-level recall while preserving the LLM's generalization capabilities.

Gaurav Pandey, Yatin Nandwani, Tahira Naseem et al. · ibm-research

Distribution matching methods for language model alignment such as Generation with Distributional Control (GDC) and Distributional Policy Gradient (DPG) have not received the same level of attention in reinforcement learning from human feedback (RLHF) as contrastive methods such as Sequence Likelihood Calibration (SLiC), Direct Preference Optimization (DPO) and its variants. We identify high variance of the gradient estimate as the primary reason for the lack of success of these methods and propose a self-normalized baseline to reduce the variance. We further generalize the target distribution in DPG, GDC and DPO by using Bayes' rule to define the reward-conditioned posterior. The resulting approach, referred to as BRAIn - Bayesian Reward-conditioned Amortized Inference acts as a bridge between distribution matching methods and DPO and significantly outperforms prior art in summarization and Antropic HH tasks.

Vivek Anand, Bharat Lohani, Gaurav Pandey et al.

Autonomous vehicles (AVs) heavily rely on LiDAR perception for environment understanding and navigation. LiDAR intensity provides valuable information about the reflected laser signals and plays a crucial role in enhancing the perception capabilities of AVs. However, accurately simulating LiDAR intensity remains a challenge due to the unavailability of material properties of the objects in the environment, and complex interactions between the laser beam and the environment. The proposed method aims to improve the accuracy of intensity simulation by incorporating physics-based modalities within the deep learning framework. One of the key entities that captures the interaction between the laser beam and the objects is the angle of incidence. In this work we demonstrate that the addition of the LiDAR incidence angle as a separate input to the deep neural networks significantly enhances the results. We present a comparative study between two prominent deep learning architectures: U-NET a Convolutional Neural Network (CNN), and Pix2Pix a Generative Adversarial Network (GAN). We implemented these two architectures for the intensity prediction task and used SemanticKITTI and VoxelScape datasets for experiments. The comparative analysis reveals that both architectures benefit from the incidence angle as an additional input. Moreover, the Pix2Pix architecture outperforms U-NET, especially when the incidence angle is incorporated.

Dinesh Khandelwal, Gnana Prakash Punnavajhala, GPS Bhargav et al.

Multi-turn tool calling is essential for LLMs to function as autonomous agents, yet synthesizing the training data required for these capabilities remains a fundamental challenge. Existing synthetic data generation pipelines often produce unrealistic dialogues for two reasons: they chain tools that are only superficially compatible rather than aligned with meaningful user tasks, and they generate dialogues in one shot, which often introduces arguments that were neither provided by the user nor produced by prior tool calls. These issues also lead to a severe underrepresentation of multi-step tool interactions. We introduce ToolWeave, a structured framework for synthesizing realistic multi-turn tool-calling dialogues. ToolWeave support realistic multi-step workflows (or tool sequences) by constructing tools with built-in dependencies and filters the workflows based on alignment with user goals. It reduces parameter hallucination by using a fine-grained planning stage that explicitly tracks parameter provenance. As a result, ToolWeave-generated synthetic dialogues contain more multi-step tool interactions (45%) and fewer hallucinations in parameters and tool names. Consequently, LLMs fine-tuned on ToolWeave consistently outperform those fine-tuned on prior datasets across three public benchmarks. Notably, Llama-3.1-70B fine-tuned on ToolWeave achieves 39.75% on BFCL-V3 multi-turn, compared to 23.50% when fine-tuned on SOTA ToolFlow data.

Ximeng Tao, Pardis Taghavi, Dimitar Filev et al.

Vision-language models (VLMs) have emerged as a promising direction for end-to-end autonomous driving (AD) by jointly modeling visual observations, driving context, and language-based reasoning. However, existing VLM-based systems face a trade-off between high-level reasoning and motion planning: large models offer strong semantic understanding but are costly to adapt for precise control, whereas small VLM models can be fine-tuned efficiently but often exhibit weaker reasoning. We propose NaviDriveVLM, a decoupled framework that separates reasoning from action generation using a large-scale Navigator and a lightweight trainable Driver. This design preserves reasoning ability, reduces training cost, and provides an explicit interpretable intermediate representation for downstream planning. Experiments on the nuScenes benchmark show that NaviDriveVLM outperforms large VLM baselines in end-to-end motion planning.

Akhil Nagariya, Dimitar Filev, Srikanth Saripalli et al.

In the area of autonomous driving, navigating off-road terrains presents a unique set of challenges, from unpredictable surfaces like grass and dirt to unexpected obstacles such as bushes and puddles. In this work, we present a novel learning-based local planner that addresses these challenges by directly capturing human driving nuances from real-world demonstrations using only a monocular camera. The key features of our planner are its ability to navigate in challenging off-road environments with various terrain types and its fast learning capabilities. By utilizing minimal human demonstration data (5-10 mins), it quickly learns to navigate in a wide array of off-road conditions. The local planner significantly reduces the real world data required to learn human driving preferences. This allows the planner to apply learned behaviors to real-world scenarios without the need for manual fine-tuning, demonstrating quick adjustment and adaptability in off-road autonomous driving technology.

Aviad Susman, Rupak Krishnamurthy, Yan Chak Li et al.

Effectively modeling multimodal longitudinal data is a pressing need in various application areas, especially biomedicine. Despite this, few approaches exist in the literature for this problem, with most not adequately taking into account the multimodality of the data. In this study, we developed multiple configurations of a novel multimodal and longitudinal learning framework, Longitudinal Ensemble Integration (LEI), for sequential classification. We evaluated LEI's performance, and compared it against existing approaches, for the early detection of dementia, which is among the most studied multimodal sequential classification tasks. LEI outperformed these approaches due to its use of intermediate base predictions arising from the individual data modalities, which enabled their better integration over time. LEI's design also enabled the identification of features that were consistently important across time for the effective prediction of dementia-related diagnoses. Overall, our work demonstrates the potential of LEI for sequential classification from longitudinal multimodal data.

Dhruvesh Patel, Benjamin Rozonoyer, Gaurav Pandey et al.

In many domains generating variable length sequences through insertions provides greater flexibility over autoregressive models. However, the action space of insertion models is much larger than that of autoregressive models (ARMs) making the learning challenging. To address this, we incorporate trainable order dynamics into the target rates for discrete flow matching, and show that with suitable choices of parameterizations, joint training of the target order dynamics and the generator is tractable without the need for numerical simulation. As the generative insertion model, we use a variable length masked diffusion model, which generates by inserting and filling mask tokens. On graph traversal tasks for which a locally optimal insertion order is known, we explore the choices of parameterization empirically and demonstrate the trade-offs between flexibility, training stability and generation quality. On de novo small molecule generation, we find that the learned order dynamics leads to an increase in the number of valid molecules generated and improved quality, when compared to uniform order dynamics.

Ximeng Tao, Dimitar Filev, Gaurav Pandey

Contrastive Language-Image Pre-training (CLIP) has shown impressive performance in aligning visual and textual representations. Recent studies have extended this paradigm to 3D vision to improve scene understanding for autonomous driving. A common strategy is to employ pairwise cosine similarity between modalities to guide the training of a 3D encoder. However, considering the similarity between individual modality pairs rather than all modalities jointly fails to ensure consistent and unified alignment across the entire multimodal space. In this paper, we propose a Contrastive Tensor Pre-training (CTP) framework that simultaneously aligns multiple modalities in a unified embedding space to enhance end-to-end autonomous driving. Compared with pairwise cosine similarity alignment, our method extends the 2D similarity matrix into a multimodal similarity tensor. Furthermore, we introduce a tensor loss to enable joint contrastive learning across all modalities. For experimental validation of our framework, we construct a text-image-point cloud triplet dataset derived from existing autonomous driving datasets. The results show that our proposed unified multimodal alignment framework achieves favorable performance for both scenarios: (i) aligning a 3D encoder with pretrained CLIP encoders, and (ii) pretraining all encoders from scratch.

Mayank Mishra, Dhiraj Madan, Gaurav Pandey et al.

Recent methods for knowledge grounded dialogs generate responses by incorporating information from an external textual document. These methods do not require the exact document to be known during training and rely on the use of a retrieval system to fetch relevant documents from a large index. The documents used to generate the responses are modeled as latent variables whose prior probabilities need to be estimated. Models such as RAG and REALM, marginalize the document probabilities over the documents retrieved from the index to define the log likelihood loss function which is optimized end-to-end. In this paper, we develop a variational approach to the above technique wherein, we instead maximize the Evidence Lower bound (ELBO). Using a collection of three publicly available open-conversation datasets, we demonstrate how the posterior distribution, that has information from the ground-truth response, allows for a better approximation of the objective function during training. To overcome the challenges associated with sampling over a large knowledge collection, we develop an efficient approach to approximate the ELBO. To the best of our knowledge we are the first to apply variational training for open-scale unsupervised knowledge grounded dialog systems.

Subodh Mishra, Armin Parchami, Enrique Corona et al.

This work presents a technique for localization of a smart infrastructure node, consisting of a fisheye camera, in a prior map. These cameras can detect objects that are outside the line of sight of the autonomous vehicles (AV) and send that information to AVs using V2X technology. However, in order for this information to be of any use to the AV, the detected objects should be provided in the reference frame of the prior map that the AV uses for its own navigation. Therefore, it is important to know the accurate pose of the infrastructure camera with respect to the prior map. Here we propose to solve this localization problem in two steps, \textit{(i)} we perform feature matching between perspective projection of fisheye image and bird's eye view (BEV) satellite imagery from the prior map to estimate an initial camera pose, \textit{(ii)} we refine the initialization by maximizing the Mutual Information (MI) between intensity of pixel values of fisheye image and reflectivity of 3D LiDAR points in the map data. We validate our method on simulated data and also present results with real world data.

Elijah S. Lee, Ankit Vora, Armin Parchami et al.

Vehicle localization is essential for autonomous vehicle (AV) navigation and Advanced Driver Assistance Systems (ADAS). Accurate vehicle localization is often achieved via expensive inertial navigation systems or by employing compute-intensive vision processing (LiDAR/camera) to augment the low-cost and noisy inertial sensors. Here we have developed a framework for fusing the information obtained from a smart infrastructure node (ix-node) with the autonomous vehicles on-board localization engine to estimate the robust and accurate pose of the ego-vehicle even with cheap inertial sensors. A smart ix-node is typically used to augment the perception capability of an autonomous vehicle, especially when the onboard perception sensors of AVs are blocked by the dynamic and static objects in the environment thereby making them ineffectual. In this work, we utilize this perception output from an ix-node to increase the localization accuracy of the AV. The fusion of ix-node perception output with the vehicle's low-cost inertial sensors allows us to perform reliable vehicle localization without the need for relying on expensive inertial navigation systems or compute-intensive vision processing onboard the AVs. The proposed approach has been tested on real-world datasets collected from a test track in Ann Arbor, Michigan. Detailed analysis of the experimental results shows that incorporating ix-node data improves localization performance.

Ashish Kumar, James R. McBride, Gaurav Pandey

We propose an end-to-end real time framework to generate high resolution graphics grade textured 3D map of urban environment. The generated detailed map finds its application in the precise localization and navigation of autonomous vehicles. It can also serve as a virtual test bed for various vision and planning algorithms as well as a background map in the computer games. In this paper, we focus on two important issues: (i) incrementally generating a map with coherent 3D surface, in real time and (ii) preserving the quality of color texture. To handle the above issues, firstly, we perform a pose-refinement procedure which leverages camera image information, Delaunay triangulation and existing scan matching techniques to produce high resolution 3D map from the sparse input LIDAR scan. This 3D map is then texturized and accumulated by using a novel technique of ray-filtering which handles occlusion and inconsistencies in pose-refinement. Further, inspired by human fovea, we introduce foveal-processing which significantly reduces the computation time and also assists ray-filtering to maintain consistency in color texture and coherency in 3D surface of the output map. Moreover, we also introduce texture error (TE) and mean texture mapping error (MTME), which provides quantitative measure of texturing and overall quality of the textured maps.

Biswesh Mohapatra, Gaurav Pandey, Danish Contractor et al.

Popular dialog datasets such as MultiWOZ are created by providing crowd workers an instruction, expressed in natural language, that describes the task to be accomplished. Crowd workers play the role of a user and an agent to generate dialogs to accomplish tasks involving booking restaurant tables, calling a taxi etc. In this paper, we present a data creation strategy that uses the pre-trained language model, GPT2, to simulate the interaction between crowd workers by creating a user bot and an agent bot. We train the simulators using a smaller percentage of actual crowd-generated conversations and their corresponding instructions. We demonstrate that by using the simulated data, we achieve significant improvements in low-resource settings on two publicly available datasets - the MultiWOZ dataset and the Persona chat dataset.

Sarah E. Houts, Nahid Pervez, Umair Ibrahim et al.

There is a growing need for vehicle positioning information to support Advanced Driver Assistance Systems (ADAS), Connectivity (V2X), and Autonomous Driving (AD) features. These range from a need for road determination ($<$5 meters), lane determination ($<$1.5 meters), and determining where the vehicle is within the lane ($<$0.3 meters). This paper presents the Ford Highway Driving RTK (Ford-HDR) dataset. This dataset includes nearly 30,000 km of data collected primarily on North American highways during a driving campaign designed to validate driver assistance features in 2018. This includes data from a representative automotive production GNSS used primarily for turn-by-turn navigation as well as an Inertial Navigation System (INS) which couples two survey-grade GNSS receivers with a tactical grade Inertial Measurement Unit (IMU) to act as ground truth. The latter utilized networked Real-Time Kinematic (RTK) GNSS corrections delivered over a cellular modem in real-time. This dataset is being released into the public domain to spark further research in the community.

Lior Lukov, Sanjay Chawla, Wei Liu et al.

In this paper, we will show that the recently introduced graphical model: Conditional Random Fields (CRF) provides a template to integrate micro-level information about biological entities into a mathematical model to understand their macro-level behavior. More specifically, we will apply the CRF model to an important classification problem in protein science, namely the secondary structure prediction of proteins based on the observed primary structure. A comparison on benchmark data sets against twenty-eight other methods shows that not only does the CRF model lead to extremely accurate predictions but the modular nature of the model and the freedom to integrate disparate, overlapping and non-independent sources of information, makes the model an extremely versatile tool to potentially solve many other problems in bioinformatics.

Subodh Mishra, Philip R. Osteen, Gaurav Pandey et al.

In this paper we perform an experimental comparison of three different target based 3D-LIDAR camera calibration algorithms. We briefly elucidate the mathematical background behind each method and provide insights into practical aspects like ease of data collection for all of them. We extensively evaluate these algorithms on a sensor suite which consists multiple cameras and LIDARs by assessing their robustness to random initialization and by using metrics like Mean Line Re-projection Error (MLRE) and Factory Stereo Calibration Error. We also show the effect of noisy sensor on the calibration result from all the algorithms and conclude with a note on which calibration algorithm should be used under what circumstances.

Ankit Vora, Siddharth Agarwal, Gaurav Pandey et al.

This paper presents a localization technique using aerial imagery maps and LIDAR based ground reflectivity for autonomous vehicles in urban environments. Traditional localization techniques using LIDAR reflectivity rely on high definition reflectivity maps generated from a mapping vehicle. The cost and effort required to maintain such prior maps are generally very high because it requires a fleet of expensive mapping vehicles. In this work we propose a localization technique where the vehicle localizes using aerial/satellite imagery, eradicating the need to develop and maintain complex high-definition maps. The proposed technique has been tested on a real world dataset collected from a test track in Ann Arbor, Michigan. This research concludes that aerial imagery based maps provides real-time localization performance similar to state-of-the-art LIDAR based maps for autonomous vehicles in urban environments at reduced costs.

Subodh Mishra, Gaurav Pandey, Srikanth Saripalli

This work presents an extrinsic parameter estimation algorithm between a 3D LIDAR and a Projective Camera using a marker-less planar target, by exploiting Planar Surface Point to Plane and Planar Edge Point to back-projected Plane geometric constraints. The proposed method uses the data collected by placing the planar board at different poses in the common field of view of the LIDAR and the Camera. The steps include, detection of the target and the edges of the target in LIDAR and Camera frames, matching the detected planes and lines across both the sensing modalities and finally solving a cost function formed by the aforementioned geometric constraints that link the features detected in both the LIDAR and the Camera using non-linear least squares. We have extensively validated our algorithm using two Basler Cameras, Velodyne VLP-32 and Ouster OS1 LIDARs.

Gaurav Pandey, Dinesh Raghu, Sachindra Joshi

Sequence to sequence models attempt to capture the correlation between all the words in the input and output sequences. While this is quite useful for machine translation where the correlation among the words is indeed quite strong, it becomes problematic for conversation modelling where the correlation is often at a much abstract level. In contrast, humans tend to focus on the essential concepts discussed in the conversation context and generate responses accordingly. In this paper, we attempt to mimic this response generating mechanism by learning the essential concepts in the context and response in an unsupervised manner. The proposed model, referred to as Mask \& Focus maps the input context to a sequence of concepts which are then used to generate the response concepts. Together, the context and the response concepts generate the final response. In order to learn context concepts from the training data automatically, we \emph{mask} words in the input and observe the effect of masking on response generation. We train our model to learn those response concepts that have high mutual information with respect to the context concepts, thereby guiding the model to \emph{focus} on the context concepts. Mask \& Focus achieves significant improvement over the existing baselines in several established metrics for dialogues.

Tom Roussel, Punarjay Chakravarty, Gaurav Pandey et al.

We describe a Deep-Geometric Localizer that is able to estimate the full 6 Degree of Freedom (DoF) global pose of the camera from a single image in a previously mapped environment. Our map is a topo-metric one, with discrete topological nodes whose 6 DoF poses are known. Each topo-node in our map also comprises of a set of points, whose 2D features and 3D locations are stored as part of the mapping process. For the mapping phase, we utilise a stereo camera and a regular stereo visual SLAM pipeline. During the localization phase, we take a single camera image, localize it to a topological node using Deep Learning, and use a geometric algorithm (PnP) on the matched 2D features (and their 3D positions in the topo map) to determine the full 6 DoF globally consistent pose of the camera. Our method divorces the mapping and the localization algorithms and sensors (stereo and mono), and allows accurate 6 DoF pose estimation in a previously mapped environment using a single camera. With potential VR/AR and localization applications in single camera devices such as mobile phones and drones, our hybrid algorithm compares favourably with the fully Deep-Learning based Pose-Net that regresses pose from a single image in simulated as well as real environments.

Tyler G. R. Reid, Nahid Pervez, Umair Ibrahim et al.

There is a growing need for vehicle positioning information to support Advanced Driver Assistance Systems (ADAS), Connectivity (V2X), and Automated Driving (AD) features. These range from a need for road determination (<5 meters), lane determination (<1.5 meters), and determining where the vehicle is within the lane (<0.3 meters). This work examines the performance of Global Navigation Satellite Systems (GNSS) on 30,000 km of North American highways to better understand the automotive positioning needs it meets today and what might be possible in the near future with wide area GNSS correction services and multi-frequency receivers. This includes data from a representative automotive production GNSS used primarily for turn-by-turn navigation as well as an Inertial Navigation System which couples two survey grade GNSS receivers with a tactical grade Inertial Measurement Unit (IMU) to act as ground truth. The latter utilized networked Real-Time Kinematic (RTK) GNSS corrections delivered over a cellular modem in real-time. We assess on-road GNSS accuracy, availability, and continuity. Availability and continuity are broken down in terms of satellite visibility, satellite geometry, position type (RTK fixed, RTK float, or standard positioning), and RTK correction latency over the network. Results show that current automotive solutions are best suited to meet road determination requirements at 98% availability but are less suitable for lane determination at 57%. Multi-frequency receivers with RTK corrections were found more capable with road determination at 99.5%, lane determination at 98%, and highway-level lane departure protection at 91%.

Tyler G. R. Reid, Sarah E. Houts, Robert Cammarata et al.

Autonomous vehicles require precise knowledge of their position and orientation in all weather and traffic conditions for path planning, perception, control, and general safe operation. Here we derive these requirements for autonomous vehicles based on first principles. We begin with the safety integrity level, defining the allowable probability of failure per hour of operation based on desired improvements on road safety today. This draws comparisons with the localization integrity levels required in aviation and rail where similar numbers are derived at 10^-8 probability of failure per hour of operation. We then define the geometry of the problem, where the aim is to maintain knowledge that the vehicle is within its lane and to determine what road level it is on. Longitudinal, lateral, and vertical localization error bounds (alert limits) and 95% accuracy requirements are derived based on US road geometry standards (lane width, curvature, and vertical clearance) and allowable vehicle dimensions. For passenger vehicles operating on freeway roads, the result is a required lateral error bound of 0.57 m (0.20 m, 95%), a longitudinal bound of 1.40 m (0.48 m, 95%), a vertical bound of 1.30 m (0.43 m, 95%), and an attitude bound in each direction of 1.50 deg (0.51 deg, 95%). On local streets, the road geometry makes requirements more stringent where lateral and longitudinal error bounds of 0.29 m (0.10 m, 95%) are needed with an orientation requirement of 0.50 deg (0.17 deg, 95%).

Rohith AP, Ambedkar Dukkipati, Gaurav Pandey

The primary objective of domain adaptation methods is to transfer knowledge from a source domain to a target domain that has similar but different data distributions. Thus, in order to correctly classify the unlabeled target domain samples, the standard approach is to learn a common representation for both source and target domain, thereby indirectly addressing the problem of learning a classifier in the target domain. However, such an approach does not address the task of classification in the target domain directly. In contrast, we propose an approach that directly addresses the problem of learning a classifier in the unlabeled target domain. In particular, we train a classifier to correctly classify the training samples while simultaneously classifying the samples in the target domain in an unsupervised manner. The corresponding model is referred to as Discriminative Encoding for Domain Adaptation (DEDA). We show that this simple approach for performing unsupervised domain adaptation is indeed quite powerful. Our method achieves state of the art results in unsupervised adaptation tasks on various image classification benchmarks. We also obtained state of the art performance on domain adaptation in Amazon reviews sentiment classification dataset. We perform additional experiments when the source data has less labeled examples and also on zero-shot domain adaptation task where no target domain samples are used for training.

Dhiraj Madan, Dinesh Raghu, Gaurav Pandey et al.

Recently several deep learning based models have been proposed for end-to-end learning of dialogs. While these models can be trained from data without the need for any additional annotations, it is hard to interpret them. On the other hand, there exist traditional state based dialog systems, where the states of the dialog are discrete and hence easy to interpret. However these states need to be handcrafted and annotated in the data. To achieve the best of both worlds, we propose Latent State Tracking Network (LSTN) using which we learn an interpretable model in unsupervised manner. The model defines a discrete latent variable at each turn of the conversation which can take a finite set of values. Since these discrete variables are not present in the training data, we use EM algorithm to train our model in unsupervised manner. In the experiments, we show that LSTN can help achieve interpretability in dialog models without much decrease in performance compared to end-to-end approaches.

Ana Stanescu, Gaurav Pandey

Heterogeneous ensembles built from the predictions of a wide variety and large number of diverse base predictors represent a potent approach to building predictive models for problems where the ideal base/individual predictor may not be obvious. Ensemble selection is an especially promising approach here, not only for improving prediction performance, but also because of its ability to select a collectively predictive subset, often a relatively small one, of the base predictors. In this paper, we present a set of algorithms that explicitly incorporate ensemble diversity, a known factor influencing predictive performance of ensembles, into a reinforcement learning framework for ensemble selection. We rigorously tested these approaches on several challenging problems and associated data sets, yielding that several of them produced more accurate ensembles than those that don't explicitly consider diversity. More importantly, these diversity-incorporating ensembles were much smaller in size, i.e., more parsimonious, than the latter types of ensembles. This can eventually aid the interpretation or reverse engineering of predictive models assimilated into the resultant ensemble(s).

Juan Castorena, Gint Puskorius, Gaurav Pandey

This work proposes a novel motion guided method for target-less self-calibration of a LiDAR and camera and use the re-projection of LiDAR points onto the image reference frame for real-time depth upsampling. The calibration parameters are estimated by optimizing an objective function that penalizes distances between 2D and re-projected 3D motion vectors obtained from time-synchronized image and point cloud sequences. For upsampling, a simple, yet effective and time efficient formulation that minimizes depth gradients subject to an equality constraint involving the LiDAR measurements is proposed. Validation is performed on recorded real data from urban environments and demonstrations that our two methods are effective and suitable to mobile robotics and autonomous vehicle applications imposing real-time requirements is shown.

Unnat Jain, Vinay P. Namboodiri, Gaurav Pandey

Robust visual place recognition (VPR) requires scene representations that are invariant to various environmental challenges such as seasonal changes and variations due to ambient lighting conditions during day and night. Moreover, a practical VPR system necessitates compact representations of environmental features. To satisfy these requirements, in this paper we suggest a modification to the existing pipeline of VPR systems to incorporate supervised hashing. The modified system learns (in a supervised setting) compact binary codes from image feature descriptors. These binary codes imbibe robustness to the visual variations exposed to it during the training phase, thereby, making the system adaptive to severe environmental changes. Also, incorporating supervised hashing makes VPR computationally more efficient and easy to implement on simple hardware. This is because binary embeddings can be learned over simple-to-compute features and the distance computation is also in the low-dimensional hamming space of binary codes. We have performed experiments on several challenging data sets covering seasonal, illumination and viewpoint variations. We also compare two widely used supervised hashing methods of CCAITQ and MLH and show that this new pipeline out-performs or closely matches the state-of-the-art deep learning VPR methods that are based on high-dimensional features extracted from pre-trained deep convolutional neural networks.

Nikhil Mehta, James R. McBride, Gaurav Pandey

This paper presents a mutual information (MI) based algorithm for the estimation of full 6-degree-of-freedom (DOF) rigid body transformation between two overlapping point clouds. We first divide the scene into a 3D voxel grid and define simple to compute features for each voxel in the scan. The two scans that need to be aligned are considered as a collection of these features and the MI between these voxelized features is maximized to obtain the correct alignment of scans. We have implemented our method with various simple point cloud features (such as number of points in voxel, variance of z-height in voxel) and compared the performance of the proposed method with existing point-to-point and point-to- distribution registration methods. We show that our approach has an efficient and fast parallel implementation on GPU, and evaluate the robustness and speed of the proposed algorithm on two real-world datasets which have variety of dynamic scenes from different environments.

Gaurav Pandey, Ambedkar Dukkipati

In recent years, deep discriminative models have achieved extraordinary performance on supervised learning tasks, significantly outperforming their generative counterparts. However, their success relies on the presence of a large amount of labeled data. How can one use the same discriminative models for learning useful features in the absence of labels? We address this question in this paper, by jointly modeling the distribution of data and latent features in a manner that explicitly assigns zero probability to unobserved data. Rather than maximizing the marginal probability of observed data, we maximize the joint probability of the data and the latent features using a two step EM-like procedure. To prevent the model from overfitting to our initial selection of latent features, we use adversarial regularization. Depending on the task, we allow the latent features to be one-hot or real-valued vectors and define a suitable prior on the features. For instance, one-hot features correspond to class labels and are directly used for the unsupervised and semi-supervised classification task, whereas real-valued feature vectors are fed as input to simple classifiers for auxiliary supervised discrimination tasks. The proposed model, which we dub discriminative encoder (or DisCoder), is flexible in the type of latent features that it can capture. The proposed model achieves state-of-the-art performance on several challenging tasks.

Gaurav Pandey, Ambedkar Dukkipati

Deep convolutional networks have achieved the state-of-the-art for semantic image segmentation tasks. However, training these networks requires access to densely labeled images, which are known to be very expensive to obtain. On the other hand, the web provides an almost unlimited source of images annotated at the image level. How can one utilize this much larger weakly annotated set for tasks that require dense labeling? Prior work often relied on localization cues, such as saliency maps, objectness priors, bounding boxes etc., to address this challenging problem. In this paper, we propose a model that generates auxiliary labels for each image, while simultaneously forcing the output of the CNN to satisfy the mean-field constraints imposed by a conditional random field. We show that one can enforce the CRF constraints by forcing the distribution at each pixel to be close to the distribution of its neighbors. This is in stark contrast with methods that compute a recursive expansion of the mean-field distribution using a recurrent architecture and train the resultant distribution. Instead, the proposed model adds an extra loss term to the output of the CNN, and hence, is faster than recursive implementations. We achieve the state-of-the-art for weakly supervised semantic image segmentation on VOC 2012 dataset, assuming no manually labeled pixel level information is available. Furthermore, the incorporation of conditional random fields in CNN incurs little extra time during training.

Gaurav Pandey, Ambedkar Dukkipati

We propose a neural network based approach for learning topics from text and image datasets. The model makes no assumptions about the conditional distribution of the observed features given the latent topics. This allows us to perform topic modelling efficiently using sentences of documents and patches of images as observed features, rather than limiting ourselves to words. Moreover, the proposed approach is online, and hence can be used for streaming data. Furthermore, since the approach utilizes neural networks, it can be implemented on GPU with ease, and hence it is very scalable.

Gaurav Pandey, Ambedkar Dukkipati

In this paper, we address the problem of conditional modality learning, whereby one is interested in generating one modality given the other. While it is straightforward to learn a joint distribution over multiple modalities using a deep multimodal architecture, we observe that such models aren't very effective at conditional generation. Hence, we address the problem by learning conditional distributions between the modalities. We use variational methods for maximizing the corresponding conditional log-likelihood. The resultant deep model, which we refer to as conditional multimodal autoencoder (CMMA), forces the latent representation obtained from a single modality alone to be `close' to the joint representation obtained from multiple modalities. We use the proposed model to generate faces from attributes. We show that the faces generated from attributes using the proposed model, are qualitatively and quantitatively more representative of the attributes from which they were generated, than those obtained by other deep generative models. We also propose a secondary task, whereby the existing faces are modified by modifying the corresponding attributes. We observe that the modifications in face introduced by the proposed model are representative of the corresponding modifications in attributes.

Gaurav Pandey, Ambedkar Dukkipati

The aim of the paper is to provide an exact approach for generating a Poisson process sampled from a hierarchical CRM, without having to instantiate the infinitely many atoms of the random measures. We use completely random measures~(CRM) and hierarchical CRM to define a prior for Poisson processes. We derive the marginal distribution of the resultant point process, when the underlying CRM is marginalized out. Using well known properties unique to Poisson processes, we were able to derive an exact approach for instantiating a Poisson process with a hierarchical CRM prior. Furthermore, we derive Gibbs sampling strategies for hierarchical CRM models based on Chinese restaurant franchise sampling scheme. As an example, we present the sum of generalized gamma process (SGGP), and show its application in topic-modelling. We show that one can determine the power-law behaviour of the topics and words in a Bayesian fashion, by defining a prior on the parameters of SGGP.

Gaurav Pandey, Ambedkar Dukkipati

To achieve acceptable performance for AI tasks, one can either use sophisticated feature extraction methods as the first layer in a two-layered supervised learning model, or learn the features directly using a deep (multi-layered) model. While the first approach is very problem-specific, the second approach has computational overheads in learning multiple layers and fine-tuning of the model. In this paper, we propose an approach called wide learning based on arc-cosine kernels, that learns a single layer of infinite width. We propose exact and inexact learning strategies for wide learning and show that wide learning with single layer outperforms single layer as well as deep architectures of finite width for some benchmark datasets.