What question did this study set out to answer?

December 1, 2015

Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture

View Full Paper

Ask AI

Bookmark

Authors

DEDavid EigenSupélec RFRob FergusSupélec

Discussion

Member takes

Implication

Demonstrates a method to predict depth, surface normals, and semantic labels using a common architecture, suggesting advancements in computer vision tasks.

Key Points

The central aim is to predict depth, surface normals, and semantic labels using a unified architecture.
Utilized a multiscale convolutional network adaptable to various tasks with minimal modifications.
Predicted outputs directly from input images without employing superpixels or low-level segmentation.
Refined predictions progressively through a sequence of scales.
Achieved state-of-the-art performance on benchmarks for depth prediction, surface normal estimation, and semantic labeling.

Cite This Study

Eigen et al. (2015) studied this question.

synapsesocial.com/papers/69d8be37d2f7327e70ae41da https://doi.org/10.1109/iccv.2015.304

View Full Paper

Ask AI

Bookmark

Implication

Demonstrates a method to predict depth, surface normals, and semantic labels using a common architecture, suggesting advancements in computer vision tasks.

Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture

Discussion

Member takes

Implication

Key Points

Cite This Study

Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider