MSR Speaking Qualifier

Title: Introducing Generative Models to Facilitate Multi-Task Visual Learning Abstract: Motivated by multi-task learning of shared feature representations, this talk considers a novel problem of learning a shared generative model that can facilitate multi-task learning. We present two systems to utilize generative modeling for other visual tasks. The first system focuses on learning a generative [...]

Title: GPU based perception via search for object pose estimation with RGB data   Abstract: Known object pose estimation is essential for a robot to interact with the real world.  It is the first and fundamental task if the robot wants to manipulate the object.  This problem is particularly challenging when the environment is complicated [...]

Title: Multi-view NRSfM: Affordable setup for high-fidelity 3D reconstruction   Abstract: Triangulating a point in 3D space should only require two corresponding camera projections. However in practice, expensive multi-view setups -- involving tens sometimes hundreds of cameras -- are required to obtain the high fidelity 3D reconstructions necessary for many modern applications. In this talk, we argue [...]

ZOOM Link: https://www.google.com/url?q=https://cmu.zoom.us/j/93845075967?pwd%3DbndGc3NvaUVDVFFTTDZvektrNWJqdz09&sa=D&source=calendar&ust=1623592142330000&usg=AOvVaw1xfNPT5c59CQGKzR2bw5sO   ID: 93845075967 Passcode: 159459 Title: 3D SLAM for Powered Lower Limb Prosthesis Abstract: During locomotion, humans use visual feedback to adjust their leg movement when navigating the environment. This natural behavior is lost, however, for lower-limb amputees, as current control strategies of prosthetic legs do not typically consider environment perception. With [...]

Title: Learnable Spatio-Temporal Map Embeddings for Deep Inertial Localization Abstract: Pedestrian localization systems often fuse inertial odometry with map information via hand-defined methods to reduce odometry drift, but such methods are sensitive to noise and struggle to generalize across odometry sources. To address the robustness problem in map utilization, we propose a system that forms a [...]

Title: Rapid Subsurface Exploration with Multiple Aerial Robots Abstract: This thesis develops a robotic exploration framework that allows for rapid and communication-efficient mapping of unknown environments with a team of aerial robots. Aerial robots can provide rapid and agile mobility in diverse environments where ground mobility is either severely constrained or impossible. However, high-speed flight [...]

Title: BusEdge: Efficient Live Video Analytics for Transit Buses via Edge Computing Abstract: Many vehicles like transit buses are now routinely fitted with cameras. These live visual data are invaluable to achieve real-time traffic monitoring, but it is intractable to handle such a gigantic amount of data either locally or in the cloud due to [...]

Title: Multi-Echo 3D Object Detection Abstract: LiDAR sensors can be used to obtain a wide range of measurement signals other than a simple 3D point cloud, and those signals can be leveraged to improve perception tasks like 3D object detection. A single laser pulse can be partially reflected by multiple objects along its path, resulting [...]

Title: Exemplar free video retrieval.   Abstract: Video retrieval of activities has a wide range of applications. In the traditional mode of operation,  a collection of example videos describing the activities are given and the retrieval technique identifies other samples in a dataset that semantically match the examples provided. However, retrieval using a collection of example [...]

Title: Data-efficient Behavior Prediction for Safe Human-Robot Collaboration.   Abstract: Predicting human behavior is critical to facilitate safe and efficient human-robot collaboration (HRC). However, human behavior is difficult to predict due to the scarcity of human motion data. This work explores using online adaptation, an online approach, and data augmentation, an offline approach, to deal with the [...]