Learning hierarchical invariant spatio-temporal features for action recognition with independent subspace analysis

Le, Quoc V.; Zou, Will Y.; Yeung, Serena; Ng, Andrew Y.

doi:10.1109/cvpr.2011.5995496

Cited by 783 publications

(647 citation statements)

References 28 publications

Supporting

Mentioning

637

Contrasting

Unclassified

Order By: Relevance

“…It is approximately 7% better than the current best published results. Our method obtains better results compared to [3], [11]. Our method also gives better results compared to our baseline method since our camera position features capture the camera position variation for this complex and challenging dataset.…”

Section: Resultsmentioning

confidence: 75%

See 1 more Smart Citation

Modeling Interactions between Low-Level and High-Level Features for Human Action Recognition

Zhou

Wang

Xiao

et al. 2013

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYRecognizing human action in complex scenes is a challenging problem in computer vision. Some action-unrelated concepts, such as camera position features, could significantly affect the appearance of local spatio-temporal features, and therefore the performance of low-level features based methods degrades. In this letter, we define the actionunrelated concept: the position of camera as high-level features. We observe that they can serve as a prior to local spatio-temporal features for human action recognition. We encode this prior by modeling interactions between spatio-temporal features and camera position features. We infer camera position features from local spatio-temporal features via these interactions. The parameters of this model are estimated by a new maxmargin algorithm. We evaluate the proposed method on KTH, IXMAS and Youtube actions datasets. Experimental results show the effectiveness of the proposed method.

show abstract

Section: Resultsmentioning

confidence: 75%

“…Sparse local spatio-temporal (ST) features have shown advantages in human action recognition [1], [2]. Besides these hand-designed features, hierarchical invariant ST features [3] that are learned from data directly are also proposed for human action recognition. They achieve impressive performance on many realistic video datasets.…”

Section: Introductionmentioning

confidence: 99%

Modeling Interactions between Low-Level and High-Level Features for Human Action Recognition

Zhou

Wang

Xiao

et al. 2013

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Recently, there has been a growing interest in learning visual features using biologically-inspired networks, such as, Independent Component Analysis (ICA) [11] and Independent Subspace Analysis [12]. In [13], Le shows that using their learned 3D (Spatiotemporal) filters by ISA, the action recognition performance is comparable to other hand- Fig. 2: Framework of the proposed method.…”

Section: Related Workmentioning

confidence: 62%

Complex Events Detection Using Data-Driven Concepts

Yang

Shah

2012

Computer Vision – ECCV 2012

View full text Add to dashboard Cite

Abstract. Automatic event detection in a large collection of unconstrained videos is a challenging and important task. The key issue is to describe long complex video with high level semantic descriptors, which should find the regularity of events in the same category while distinguish those from different categories. This paper proposes a novel unsupervised approach to discover data-driven concepts from multi-modality signals (audio, scene and motion) to describe high level semantics of videos. Our methods consists of three main components: we first learn the low-level features separately from three modalities. Secondly we discover the datadriven concepts based on the statistics of learned features mapped to a low dimensional space using deep belief nets (DBNs). Finally, a compact and robust sparse representation is learned to jointly model the concepts from all three modalities. Extensive experimental results on large in-thewild dataset show that our proposed method significantly outperforms state-of-the-art methods.

show abstract

“…Our autoencoder model draws inspiration from Independent Subspace Analysis (ISA) [25] which was proposed for learning motion invariance. Unlike conventional sparse autoencoders which enforce sparsity in the hidden layer, the proposed autoencoder performs subspace pooling on the hidden layer activations and enforces sparsity in the pooling layer, in a way identical to ISA.…”

Section: Autoencoder With Temporal Slowness Constraintmentioning

confidence: 99%

Self-taught learning of a deep invariant representation for visual tracking via temporal slowness principle

Kuen

Lim

Lee

2015

Pattern Recognition

View full text Add to dashboard Cite

Visual representation is crucial for a visual tracking method's performances. Conventionally, visual representations adopted in visual tracking rely on hand-crafted computer vision descriptors. These descriptors were developed generically without considering tracking-specific information. In this paper, we propose to learn complex-valued invariant representations from tracked sequential image patches, via strong temporal slowness constraint and stacked convolutional autoencoders. The deep slow local representations are learned offline on unlabeled data and transferred to the observational model of our proposed tracker. The proposed observational model retains old training samples to alleviate drift, and collect negative samples which are coherent with target's motion pattern for better discriminative tracking. With the learned representation and online training samples, a logistic regression classifier is adopted to distinguish target from background, and retrained online to adapt to appearance changes. Subsequently, the observational model is integrated into a particle filter framework to peform visual tracking. Experimental results on various challenging benchmark sequences demonstrate that the proposed tracker performs favourably against several state-of-the-art trackers.

show abstract

Learning hierarchical invariant spatio-temporal features for action recognition with independent subspace analysis

Cited by 783 publications

References 28 publications

Modeling Interactions between Low-Level and High-Level Features for Human Action Recognition

Modeling Interactions between Low-Level and High-Level Features for Human Action Recognition

Complex Events Detection Using Data-Driven Concepts

Self-taught learning of a deep invariant representation for visual tracking via temporal slowness principle

Contact Info

Product

Resources

About