Learning to represent scenes with geometry and semantics
<p>Scene representation is the process of converting sensory observations of an environment into compact descriptions. Such intelligent behavior is a cornerstone of artificial intelligence. Scientists have long sought to reproduce the extraordinary ability of humans for understanding the physi...
Auteur principal: | Wang, B |
---|---|
Autres auteurs: | Trigoni, A |
Format: | Thèse |
Langue: | English |
Publié: |
2022
|
Sujets: |
Documents similaires
-
Applications of AI, machine vision and robotics /
par: Boyer, Kim L., et autres
Publié: (1995) -
Pixel-level scene understanding with deep structured models
par: Arnab, A
Publié: (2019) -
Geometry‐preserved image editing
par: Taeeun Kwon, et autres
Publié: (2024-09-01) -
Holistic, instance-level scene parsing with deep neural networks
par: Li, Q
Publié: (2020) -
Unifying motion segmentation, estimation, and tracking for complex dynamic scenes
par: Judd, K
Publié: (2019)