Zobrazeno 1 - 4
of 4
pro vyhledávání: '"Ben-Avraham, Elad"'
Autor:
Herzig, Roei, Abramovich, Ofir, Ben-Avraham, Elad, Arbelle, Assaf, Karlinsky, Leonid, Shamir, Ariel, Darrell, Trevor, Globerson, Amir
Action recognition models have achieved impressive results by incorporating scene-level annotations, such as objects, their relations, 3D structure, and more. However, obtaining annotations of scene structure for videos requires a significant amount
Externí odkaz:
http://arxiv.org/abs/2212.04821
Autor:
Ben-Avraham, Elad, Herzig, Roei, Mangalam, Karttikeya, Bar, Amir, Rohrbach, Anna, Karlinsky, Leonid, Darrell, Trevor, Globerson, Amir
This technical report describes the SViT approach for the Ego4D Point of No Return (PNR) Temporal Localization Challenge. We propose a learning framework StructureViT (SViT for short), which demonstrates how utilizing the structure of a small number
Externí odkaz:
http://arxiv.org/abs/2206.07689
Autor:
Ben-Avraham, Elad, Herzig, Roei, Mangalam, Karttikeya, Bar, Amir, Rohrbach, Anna, Karlinsky, Leonid, Darrell, Trevor, Globerson, Amir
Recent action recognition models have achieved impressive results by integrating objects, their locations and interactions. However, obtaining dense structured annotations for each frame is tedious and time-consuming, making these methods expensive t
Externí odkaz:
http://arxiv.org/abs/2206.06346
Autor:
Herzig, Roei, Ben-Avraham, Elad, Mangalam, Karttikeya, Bar, Amir, Chechik, Gal, Rohrbach, Anna, Darrell, Trevor, Globerson, Amir
Recently, video transformers have shown great success in video understanding, exceeding CNN performance; yet existing video transformer models do not explicitly model objects, although objects can be essential for recognizing actions. In this work, w
Externí odkaz:
http://arxiv.org/abs/2110.06915