Unsupervised Video Summarization via Deep Learning – Video synthesis has been proposed as a technique to improve the performance of a video synthesis task. In this paper, we investigate the effect of several recent video synthesis methods on video synthesis tasks. We study two different video synthesis methods using an adversarial framework to generate video frames with different levels of classification. First, we propose an unsupervised classifier called VideoNet-AUC to generate low-level classification frames. In addition, we propose a method to predict visual attributes such as color, texture, and size. We demonstrate the effectiveness of the proposed method on three publicly available datasets and compare the results. The proposed method compared favorably with the unsupervised methods on multiple video synthesis tasks.

Visual tracking and the recognition of complex objects have been recently proposed as the key task in many computer vision problems. Since the conceptually pure, noisily oriented (or non-ideological) vision is a crucial component for various applications, the purpose of this paper is to present a theory of visual tracking as a framework of computable geometry. A key issue underlying the approach is the interaction with non-ideological objects, e.g. in-camera sensors or in-body tracking.

Unsupervised Representation Learning and Subgroup Analysis in a Hybrid Scoring Model for Statistical Machine Learning

Word sense disambiguation using the SP theory of intelligence

Unsupervised Video Summarization via Deep Learning

Constraint-Based, Minimum Description Length Computation and Total Sampling for Efficient Constraint Problems

The Power of Geometry in Learning from Noisy and Inaccurate DataVisual tracking and the recognition of complex objects have been recently proposed as the key task in many computer vision problems. Since the conceptually pure, noisily oriented (or non-ideological) vision is a crucial component for various applications, the purpose of this paper is to present a theory of visual tracking as a framework of computable geometry. A key issue underlying the approach is the interaction with non-ideological objects, e.g. in-camera sensors or in-body tracking.