No
Yes
View More
View Less
Working...
Close
OK
Cancel
Confirm
System Message
Delete
My Schedule
An unknown error has occurred and your request could not be completed. Please contact support.
Scheduled
Scheduled
Wait Listed
Personal Calendar
Speaking
Conference Event
Meeting
Interest
There aren't any available sessions at this time.
Conflict Found
This session is already scheduled at another time. Would you like to...
Loading...
Please enter a maximum of {0} characters.
{0} remaining of {1} character maximum.
Please enter a maximum of {0} words.
{0} remaining of {1} word maximum.
must be 50 characters or less.
must be 40 characters or less.
Session Summary
We were unable to load the map image.
This has not yet been assigned to a map.
Search Catalog
Reply
Replies ()
Search
New Post
Microblog
Microblog Thread
Post Reply
Post
Your session timed out.
This web page is not optimized for viewing on a mobile device. Visit this site in a desktop browser to access the full set of features.
2019 GTC San Jose

S9874 - Learning Viewpoint Estimators with Few Examples

Session Speakers
Session Description

Humans can predict the viewpoint of novel categories of objects from just a few views, an ability that viewpoint estimation networks can't match. These networks require thousands of labeled training examples per category and struggle with unseen ones. We'll discuss our investigation into the unexplored problem of category-level few-shot viewpoint estimation. We'll describe our design of a CNN framework called MetaView, which formulates the problem as successfully learning to detect category-specific semantic key points for viewpoint estimation with only a few examples per category. Inspired by the recent success of meta-learning on other few-shot learning visual tasks, we use meta-learning to train our algorithm. We'll discuss our experiments on two benchmark datasets, which demonstrate our method's superior performance versus zero-shot performance of current methods, and show how we fine-tune them with a few examples per novel category.


Additional Information
Computer Vision
AI/Deep Learning Research Computer Vision
General
Intermediate technical
Talk
50 minutes
Session Schedule