Shubham Jain

Bryan Bo Cao, Abrar Alali, Hansi Liu et al.

Tracking subjects in videos is one of the most widely used functions in camera-based IoT applications such as security surveillance, smart city traffic safety enhancement, vehicle to pedestrian communication and so on. In the computer vision domain, tracking is usually achieved by first detecting subjects with bounding boxes, then associating detected bounding boxes across video frames. For many IoT systems, images captured by cameras are usually sent over the network to be processed at a different site that has more powerful computing resources than edge devices. However, sending entire frames through the network causes significant bandwidth consumption that may exceed the system bandwidth constraints. To tackle this problem, we propose ViFiT, a transformer-based model that reconstructs vision bounding box trajectories from phone data (IMU and Fine Time Measurements). It leverages a transformer ability of better modeling long-term time series data. ViFiT is evaluated on Vi-Fi Dataset, a large-scale multimodal dataset in 5 diverse real world scenes, including indoor and outdoor environments. To fill the gap of proper metrics of jointly capturing the system characteristics of both tracking quality and video bandwidth reduction, we propose a novel evaluation framework dubbed Minimum Required Frames (MRF) and Minimum Required Frames Ratio (MRFR). ViFiT achieves an MRFR of 0.65 that outperforms the state-of-the-art approach for cross-modal reconstruction in LSTM Encoder-Decoder architecture X-Translator of 0.98, resulting in a high frame reduction rate as 97.76%.

Faith Johnson, Bryan Bo Cao, Kristin Dana et al.

Map representations learned by expert demonstrations have shown promising research value. However, the field of visual navigation still faces challenges due to the lack of real-world human-navigation datasets that can support efficient, supervised, representation learning of environments. We present a Landmark-Aware Visual Navigation (LAVN) dataset to allow for supervised learning of human-centric exploration policies and map building. We collect RGBD observation and human point-click pairs as a human annotator explores virtual and real-world environments with the goal of full coverage exploration of the space. The human annotators also provide distinct landmark examples along each trajectory, which we intuit will simplify the task of map or graph building and localization. These human point-clicks serve as direct supervision for waypoint prediction when learning to explore in environments. Our dataset covers a wide spectrum of scenes, including rooms in indoor environments, as well as walkways outdoors. We release our dataset with detailed documentation at https://huggingface.co/datasets/visnavdataset/lavn (DOI: 10.57967/hf/2386) and a plan for long-term preservation.

Ryan Meegan, Adam D'Souza, Bryan Bo Cao et al.

Visual navigation has emerged as a practical alternative to traditional robotic navigation pipelines that rely on detailed mapping and path planning. However, constructing and maintaining 3D maps is often computationally expensive and memory-intensive. We address the problem of visual navigation when exploration videos of a large environment are available. The videos serve as a visual reference, allowing a robot to retrace the explored trajectories without relying on metric maps. Our proposed method, YOPO-Nav (You Only Pass Once), encodes an environment into a compact spatial representation composed of interconnected local 3D Gaussian Splatting (3DGS) models. During navigation, the framework aligns the robot's current visual observation with this representation and predicts actions that guide it back toward the demonstrated trajectory. YOPO-Nav employs a hierarchical design: a visual place recognition (VPR) module provides coarse localization, while the local 3DGS models refine the goal and intermediate poses to generate control actions. To evaluate our approach, we introduce the YOPO-Campus dataset, comprising 4 hours of egocentric video and robot controller inputs from over 6 km of human-teleoperated robot trajectories. We benchmark recent visual navigation methods on trajectories from YOPO-Campus using a Clearpath Jackal robot. Experimental results show YOPO-Nav provides excellent performance in image-goal navigation for real-world scenes on a physical robot. The dataset and code will be made publicly available for visual navigation and scene representation research.

Faith Johnson, Bryan Bo Cao, Ashwin Ashok et al.

Visual navigation follows the intuition that humans can navigate without detailed maps. A common approach is interactive exploration while building a topological graph with images at nodes that can be used for planning. Recent variations learn from passive videos and can navigate using complex social and semantic cues. However, a significant number of training videos are needed, large graphs are utilized, and scenes are not unseen since odometry is utilized. We introduce a new approach to visual navigation using feudal learning, which employs a hierarchical structure consisting of a worker agent, a mid-level manager, and a high-level manager. Key to the feudal learning paradigm, agents at each level see a different aspect of the task and operate at different spatial and temporal scales. Two unique modules are developed in this framework. For the high-level manager, we learn a memory proxy map in a self supervised manner to record prior observations in a learned latent space and avoid the use of graphs and odometry. For the mid-level manager, we develop a waypoint network that outputs intermediate subgoals imitating human waypoint selection during local navigation. This waypoint network is pre-trained using a new, small set of teleoperation videos that we make publicly available, with training environments different from testing environments. The resulting feudal navigation network achieves near SOTA performance, while providing a novel no-RL, no-graph, no-odometry, no-metric map approach to the image goal navigation task.

Bryan Bo Cao, Abhinav Sharma, Manavjeet Singh et al.

Edge computing has emerged as an alternative to reduce transmission and processing delay and preserve privacy of the video streams. However, the ever-increasing complexity of Deep Neural Networks (DNNs) used in video-based applications (e.g. object detection) exerts pressure on memory-constrained edge devices. Model merging is proposed to reduce the DNNs' memory footprint by keeping only one copy of merged layers' weights in memory. In existing model merging techniques, (i) only architecturally identical layers can be shared; (ii) requires computationally expensive retraining in the cloud; (iii) assumes the availability of ground truth for retraining. The re-evaluation of a merged model's performance, however, requires a validation dataset with ground truth, typically runs at the cloud. Common metrics to guide the selection of shared layers include the size or computational cost of shared layers or representation size. We propose a new model merging scheme by sharing representations (i.e., outputs of layers) at the edge, guided by representation similarity S. We show that S is extremely highly correlated with merged model's accuracy with Pearson Correlation Coefficient |r| > 0.94 than other metrics, demonstrating that representation similarity can serve as a strong validation accuracy indicator without ground truth. We present our preliminary results of the newly proposed model merging scheme with identified challenges, demonstrating a promising research future direction.

Faith Johnson, Bryan Bo Cao, Ashwin Ashok et al.

Visual navigation takes inspiration from humans, who navigate in previously unseen environments using vision without detailed environment maps. Inspired by this, we introduce a novel no-RL, no-graph, no-odometry approach to visual navigation using feudal learning to build a three tiered agent. Key to our approach is a memory proxy map (MPM), an intermediate representation of the environment learned in a self-supervised manner by the high-level manager agent that serves as a simplified memory, approximating what the agent has seen. We demonstrate that recording observations in this learned latent space is an effective and efficient memory proxy that can remove the need for graphs and odometry in visual navigation tasks. For the mid-level manager agent, we develop a waypoint network (WayNet) that outputs intermediate subgoals, or waypoints, imitating human waypoint selection during local navigation. For the low-level worker agent, we learn a classifier over a discrete action space that avoids local obstacles and moves the agent towards the WayNet waypoint. The resulting feudal navigation network offers a novel approach with no RL, no graph, no odometry, and no metric map; all while achieving SOTA results on the image goal navigation task.

Slobodan Milanko, Alexander Launi, Shubham Jain

This paper presents our findings from a multi-year effort to detect motion events early using inertial sensors in real-world settings. We believe early event detection is the next step in advancing motion tracking, and can enable just-in-time interventions, particularly for mHealth applications. Our system targets strength training workouts in the fitness domain, where users perform well-defined movements for each exercise, while wearing an inertial sensor. We collect data for 20 exercises across 12 users over 26 months. We propose an algorithm to detect repetitions before they end, to allow a user to visualize movement derived metrics in real-time. We further develop a gamified approach to display this information to the user and encourage them to perform consistent movements. Participants in a feasibility study find the gamified feedback useful in improving their form. Our system can detect repetition events as early as 500 ms before it ends, which is 2x faster and more accurate than state-of-the-art trackers. We believe our approach will open exciting avenues for tracking, detection, and gamification for fitness frameworks.

Shubham Jain, Marco Gruteser

As smartphone rooted distractions become commonplace, the lack of compelling safety measures has led to a rise in the number of injuries to distracted walkers. Various solutions address this problem by sensing a pedestrian's walking environment. Existing camera-based approaches have been largely limited to obstacle detection and other forms of object detection. Instead, we present TerraFirma, an approach that performs material recognition on the pedestrian's walking surface. We explore, first, how well commercial off-the-shelf smartphone cameras can learn texture to distinguish among paving materials in uncontrolled outdoor urban settings. Second, we aim at identifying when a distracted user is about to enter the street, which can be used to support safety functions such as warning the user to be cautious. To this end, we gather a unique dataset of street/sidewalk imagery from a pedestrian's perspective, that spans major cities like New York, Paris, and London. We demonstrate that modern phone cameras can be enabled to distinguish materials of walking surfaces in urban areas with more than 90% accuracy, and accurately identify when pedestrians transition from sidewalk to street.