Abstract | ||
---|---|---|
Visual recognition systems mounted on autonomous moving agents face the challenge of unconstrained data, but simultaneously have the opportunity to improve their performance by moving to acquire new views at test time. In this work, we first show how a recurrent neural network-based system may be trained to perform end-to-end learning of motion policies suited for this “active recognition” setting... |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/TPAMI.2018.2840991 | IEEE Transactions on Pattern Analysis and Machine Intelligence |
Keywords | Field | DocType |
Cameras,Visualization,Recurrent neural networks,Image recognition,Task analysis,Machine vision,Pipelines | Computer vision,Categorization,Active vision,Machine vision,Task analysis,Computer science,End-to-end principle,Visualization,Recurrent neural network,Look-ahead,Artificial intelligence,Machine learning | Journal |
Volume | Issue | ISSN |
41 | 7 | 0162-8828 |
Citations | PageRank | References |
3 | 0.37 | 25 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dinesh Jayaraman | 1 | 318 | 15.69 |
Kristen Grauman | 2 | 6258 | 326.34 |