Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences

By Zhiyuan Shi, Tae-Kyun Kim. 2017

83 citations recorded by the source corpus.

Existing RNN-based approaches for action recognition from depth sequences require either skeleton joints or hand-crafted depth features as inputs. An end-to-end manner, mapping from raw depth maps to action classes, is non-trivial to design due to the fact that: 1) single channel map lacks texture thus weakens the discriminative power, 2) relatively small set of depth training data. To address these challenges, we propose to learn an RNN driven by privileged information (PI) in three-steps: An encoder is pre-trained to learn a joint embedding of depth appearance and PI (i.e. skeleton joints). The learned embedding layers are then tuned in the learning step, aiming to optimize the network by exploiting PI in a form of multi-task loss. However, exploiting PI as a secondary task provides little help to improve the performance of a primary task (i.e. classification) due to the gap between them. Finally, a bridging matrix is defined to connect two tasks by discovering latent PI in the refining step. Our PI-based classification loss maintains a consistency between latent PI and predicted distribution. The latent PI and network are iteratively estimated and updated in an expectation-maximization procedure. The proposed learning process provides greater discriminative power to model subtle depth difference, while helping avoid overfitting the scarcer training data. Our experiments show significant performance gains over state-of-the-art methods on three public benchmark datasets and our newly collected Blanket dataset.

Subfields: Artificial Intelligence, Biomedical Engineering, Computer Vision and Pattern Recognition, Model Architectures

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Biomedical Engineering

In Computer Vision and Pattern Recognition

In Model Architectures

Citations in the corpus

NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understanding (2019, background): “Speciﬁcally, many of them have been evaluated based on the preliminary version [47] of our dataset, or pre-trained on it for transfer learning for other tasks [43], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79].”
RGB-D-based Human Motion Recognition with Deep Learning: A Survey (2017, methodology): “For example, methods such as SFAM [151] and PRNN [113] have pioneered the research in this direction.”
Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates (2017, neutral): “Skeleton-based action recognition has been explored in different aspects in recent years [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57].”
First-Person Hand Action Benchmark with RGB-D Videos and 3D Hand Pose Annotations (2017, background): “In fullbody human action recognition it is known that using higher level and viewpoint invariant features such as body pose can benefit action recognition [54, 74, 79, 81, 86], although this has not yet been studied in detail for hands.”
Graph Distillation for Action Detection with Privileged Modalities (2017, methodology): “Our graph distillation method is different from prior work [18,26,33,49] in that the privileged information contains multiple modalities and that the distillation directions and weights are dynamically learned rather than being predefined by human experts.”