ICAR Research Data Repository for Knowledge Management
Multi-Stage LSTM (MS-LSTM) for Action Anticipation
CSIRO RDS Repository
View Archive Info| Field | Value | |
| Title |
Multi-Stage LSTM (MS-LSTM) for Action Anticipation
|
|
| Creator |
Mohammad Sadegh Aliakbarian
|
|
| Subject |
Knowledge Representation and Machine Learning
Computer Vision |
|
| Description |
Multi-Stage LSTM (MS-LSTM) is a deep sequence learning architecture that leverage two types of descriptors, i.e., action-aware and context-aware features to determine the action label. To do it at early stages, MS-LSTM is trained with a novel loss function that encourages correct prediction as early as possible.
|
|
| Publisher |
CSIRO
|
|
| Contributor |
Fatemeh Sadat Saleh
Mathieu Salzmann Basura Fernando Lars Petersson Lars Andersson |
|
| Date |
2018-02-22
|
|
| Type |
—
|
|
| Format |
—
|
|
| Identifier |
csiro:29514
|
|
| Language |
—
|
|
| Coverage |
—
|
|
| Rights |
—
|
|