Name
Abstract | ||
|---|---|---|
In this paper, we propose an end-to-end reasoning-decision networks (RDN) approach for robust face alignment via policy gradient. Unlike the conventional coarse-to-fine approaches which likely lead to bias prediction due to poor initialization, our approach aims to learn a policy by leveraging raw pixels to reason a subset of shape candidates, sequentially making plausible decisions to remove outl... |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1109/TPAMI.2018.2885298 | IEEE Transactions on Pattern Analysis and Machine Intelligence |
Keywords | Field | DocType |
Shape,Face,Training,Neural networks,Two dimensional displays,Computer architecture | Pattern recognition,Computer science,Local optimum,Decision networks,Markov decision process,Outlier,Bellman equation,Artificial intelligence,Pixel,Initialization,Trajectory | Journal |
Volume | Issue | ISSN |
42 | 3 | 0162-8828 |
Citations | PageRank | References |
1 | 0.38 | 30 |
Authors | ||
5 | ||