Autonomous driving and advanced driver assistance systems have made significant advances in recent years. Dataset construction for road scenes can be time consuming and cost-prohibitive, even more so for the thermal domain. The thermal domain is underrepresented in publicly available datasets compared to visible-light domains. The proposed methodology cheaply produces high-quality thermal (IR) semantic segmentation networks using multiple rounds of knowledge distillation with different training domains and no IR labels. A trained visual spectrum (RGB) is used to train thermal and visual-thermal (RGBT) students. The experiment compares the use of intermediate training domains, such as RGBT for training a thermal student. The method’s key contribution is demonstrating that the best results come from directly training the IR student with an RGB teacher, without intermediate trainings.

Many students who switch away from an engineering major cite math, specifically calculus, as one of the most influential factors in that decision. This paper presents a method to improve engineering student persistence and learning in calculus without any changes in the current calculus curriculum. The proposed method combines the integration of real world engineering problems into existing Calculus II courses and engineering mentors to help students to solve calculus problems. The implementation of this method was achieved through a pilot study administered to undergraduate engineering students enrolled in a Calculus II course at a mid-size research institution in the Midwest. The results of the pilot show that engineering problems helped the student participants to learn calculus. Students benefited from working closely with their peers and their mentors in the Calculus II course, and they felt more connected to other students in math and/or engineering.

Traditional robot controllers are not designed to produce human-like reactive motion - movements lasting tens of sample periods and requiring large accelerations. One of the major obstacles to producing reactive motions with contemporary controllers is that they rely on kinematic commands. The performance of short duration motions requiring large accelerations is dominated by the motion's dynamics; kinematic commands without an accurate dynamic model of the robot and task will lead to poor performance. Conversely, this paper presents a biologically inspired “torque command” that allows the dynamics of the motion to be communicated to the controller. The commands can be produced with only minor modifications to any existing control scheme that will not impact traditional operation. In addition, sensory input can be mapped directly to presented commands for latency sensitive tasks. The ability of the new commands to express a broad range of motions with a small number of parameters is shown experimentally. The experimental results also show that the presented torque commands can be used to learn a ball intercept task.

Many of the robust visual tracking techniques utilized by augmented reality applications rely on 3D models and information extracted from images. Models enhanced with image information make it possible to initialize tracking and detect poor registration. Unfortunately, generating 3D CAD models and registering them to image information can be a time consuming operation. Regularly the process requires multiple trips between the site being modeled and the workstation used to create the model. The system presented in this work eliminates the need for a separately generated 3D model by utilizing modern structure-from-motion techniques to extract the model and associated image information directly from an image sequence. The technique can be implemented on any handheld device instrumented with a camera and network connection. The process of creating the model requires minimal user interaction in the form of a few cues to identify planar regions on the object of interest. In addition the system selects a set of keyframes for each region to capture viewpoint based appearance changes. This work also presents a robust tracking framework to take advantage of these new edge models. Performance of both the modeling technique and the tracking system are verified on several different objects.

This paper presents a method for identifying handheld devices (e.g. smart phones and pocket PCs) to facilitate the use of these devices as tangible interfaces for desktop augmented reality systems. The proposed system leverages the ability of these handheld devices to programmatically control their backlight intensity to display a binary code. The codes produced are non-intrusive, require no specialized hardware, and can be generated with most handheld devices. This technique is shown to accurately and robustly identify up to 16 different devices in under 500 msec and is easily expandable to 256 or more devices.

We present a calibration procedure to determine the kinematic parameters of an active stereo system in a robot-centric frame of reference. Our goal was to obtain a solution of sufficient accuracy that the kinematic model information can be used to estimate scene structure given the measured motion/position of the eye and target object's image location. We formulate our problem using canonical coordinates of the rotation group. which enables a particularly simple closed form solution. Additionally, this formulation and solution provides quantitative measures of the resulting solutions. Experiments verify that the solutions are accurate, 3D structure can be estimated from the kinematic models, and the algorithm can indicate when image errors are large enough to produce unreliable results.

Addresses the problem of recovering 3D geometry using an active stereo vision system. Calibration procedures can be adapted to the active stereo configuration, however, considerable effort is required to accurately model and calibrate the kinematics to avoid poor reconstruction. In the active stereo case there will also be errors due to uncertainty in the kinematics of the system. In addition, data collection needs to be automated because active stereo requires significantly more information for calibration. We present a biologically inspired neural network trained to determine the mapping between 3D geometry and stereo image points. To train the network, we have developed a system to automatically collect accurate calibration data. We compare the reconstructed 3D geometry obtained using a kinematic model based approach with our neural network approach.