
December 17, 2021
Core Data Science at Facebook
Core Data Science at Facebook
We present a new method that views object detection as a direct set prediction problem. Our approach streamlines the detection pipeline, effectively removing the need for many hand-designed components like a non-maximum suppression procedure or anchor generation that explicitly encode our prior knowledge about the task. The main ingredients of the new framework, called DEtection TRansformer or DETR, are a set-based global loss that forces unique predictions via bipartite matching, and a transformer encoder-decoder architecture. (Read more)
December 17, 2021
Core Data Science at Facebook
December 17, 2021
Computational Social Science
December 17, 2021
Common Networks Deploys Terragraph to Serve Customers in California
December 17, 2021
ARCH: Animatable Reconstruction of Clothed Humans
December 17, 2021
World Effects Framework
December 17, 2021
Wish You Were Here: Context Aware Human Generation
December 17, 2021
What Makes Training Multi-Modal Networks Hard?
December 17, 2021
Using social media data to help measure smoke exposure
December 17, 2021
Using Data to Help Communities Recover and Rebuild