Pittsburgh, Pennsylvania, United States
Contact Info
356 followers
344 connections
About
Activity
-
Dongkyun Kim, a junior #artificialintelligence major in the Carnegie Mellon University School of Computer Science, designed the winning deep-learning…
Dongkyun Kim, a junior #artificialintelligence major in the Carnegie Mellon University School of Computer Science, designed the winning deep-learning…
Liked by Shasa Antao
-
Advancements in AI help our onboard tech better understand the driving environment. Apoorv Singh explains how Transformer Neural Networks help make…
Advancements in AI help our onboard tech better understand the driving environment. Apoorv Singh explains how Transformer Neural Networks help make…
Liked by Shasa Antao
Experience & Education
Licenses & Certifications
Courses
-
Computer Graphics
15-662
-
Computer Vision
16-720B
-
Geometry based Vision
16-887
-
Machine Learning
10-601
-
Robot Autonomy
16-662
-
Robot Localization & Mapping (SLAM)
16-833A
-
Robot Mobility on Air, Land & Sea
16-665
-
Visual Recognition and Learning
16-824
Projects
-
Experimenting with Low-Resolution Point Clouds and One-Shot Learning for 3D Object Detection
-
One of the major challenges in developing a point cloud based object detection system stems from the fact that point cloud data is irregular, unordered, and usually sparse. Another challenge in using these detection systems for novel applications is that there is a limited amount of accurately labeled class data. One-shot learning can solve the problem of having a dataset with few samples per class, and in this project, we wanted to try and extrapolate concepts from a state-of-the-art one-shot…
One of the major challenges in developing a point cloud based object detection system stems from the fact that point cloud data is irregular, unordered, and usually sparse. Another challenge in using these detection systems for novel applications is that there is a limited amount of accurately labeled class data. One-shot learning can solve the problem of having a dataset with few samples per class, and in this project, we wanted to try and extrapolate concepts from a state-of-the-art one-shot learner for 2D RGB object detection and try applying these concepts to perform object detection on 3D point cloud data. .
Other creatorsSee project -
Autonomous Driving for Adverse Perceived Terrain
-
Research suggests that thousands of people are killed, hundreds of thousands are injured, and billions of dollars are wasted, every year during harsh adverse weather events. Even in the event that a high level of autonomy is achieved for autonomous vehicles in dry and sunny conditions, it will likely not translate to any reduction in these numbers, since people will always need to travel and transport goods, no matter the state of the road conditions.
The ADAPT (Autonomous Driving for…Research suggests that thousands of people are killed, hundreds of thousands are injured, and billions of dollars are wasted, every year during harsh adverse weather events. Even in the event that a high level of autonomy is achieved for autonomous vehicles in dry and sunny conditions, it will likely not translate to any reduction in these numbers, since people will always need to travel and transport goods, no matter the state of the road conditions.
The ADAPT (Autonomous Driving for Adverse Perceived Terrain) system is designed to combat this issue. The system serves to actively detect changes in road conditions and adjust the vehicle controls to account for these changes, making transitions into wet or otherwise unfavorable road conditions a much safer endeavor. Current systems are almost exclusively reactive, meaning that they only modify the vehicle’s actions after an incident has occurred, whereas the ADAPT system aims to predict future scenarios and maximize the ability to safely transport passengers and goods.
Other creatorsSee project
More activity by Shasa
-
Exciting plans for the new CMU robotics innovation center!
Exciting plans for the new CMU robotics innovation center!
Liked by Shasa Antao
-
I'm thrilled to announce my recent promotion to the position of Senior Computer Vision Engineer, coinciding with my upcoming second anniversary here…
I'm thrilled to announce my recent promotion to the position of Senior Computer Vision Engineer, coinciding with my upcoming second anniversary here…
Liked by Shasa Antao
-
Transformer DNN architectures are all the rage, but their structure is radically different from the assumptions made by NPU accelerator designers.…
Transformer DNN architectures are all the rage, but their structure is radically different from the assumptions made by NPU accelerator designers.…
Liked by Shasa Antao
-
Check out what a day in life looks like as a Machine Learning Researcher at Motional! #autonomousdriving #machinelearning #motional
Check out what a day in life looks like as a Machine Learning Researcher at Motional! #autonomousdriving #machinelearning #motional
Liked by Shasa Antao
-
I am excited to share that I am starting a new position as Lead DevOps Engineer at Viasat Inc. #WeAreViasat
I am excited to share that I am starting a new position as Lead DevOps Engineer at Viasat Inc. #WeAreViasat
Liked by Shasa Antao
-
DAF-MIT AI Accelerator Phantom program Cohort 7 took a tour of Kessel Run this week! Brothers Dr. Ethan Sneider, DAF-MIT #AI Accelerator Phantom…
DAF-MIT AI Accelerator Phantom program Cohort 7 took a tour of Kessel Run this week! Brothers Dr. Ethan Sneider, DAF-MIT #AI Accelerator Phantom…
Liked by Shasa Antao
-
Sigma-Fusion: Probabilistic Volumetric Fusion for Dense Monocular SLAM https://lnkd.in/eU_HTryi Code: https://lnkd.in/eV6zx9zz Paper:…
Sigma-Fusion: Probabilistic Volumetric Fusion for Dense Monocular SLAM https://lnkd.in/eU_HTryi Code: https://lnkd.in/eV6zx9zz Paper:…
Liked by Shasa Antao
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More