Conventional video coding methods have been developed based on the human visual system (HVS). However, in recent years, video has occupied a huge portion of internet traffic, and the mount of video data for machine consumption has increased rapidly due to the progress of neural networks. This paper proposes a novel machine-attention-based video coding method for machines. Inspired by the saliency-driven research, we first extract attention regions, sensitively affecting the machine vision performance, from the object detection network. Subsequently, a maximum a posterior (MAP)-based bit allocation method is applied to assign more bits to the attention regions. Our proposed method helps to maintain high machine vision performance whereas reducing the bitrate. Experimental results show that our proposed method achieves up to 34.89% bjøntegaard delta (BD)-rate reduction for the video dataset and up to 44.70% BD-rate reduction for the image dataset compared to state-of-the-art video coding technology.

The existing industry evaluation method utilizes the method of collecting the structured information such as the financial information of the companies included in the relevant industry and deriving the industrial evaluation index through the statistical analysis model. This method takes a long time to calculate the structured data and cause the time delay problem. In this paper, to solve this time delay problem, we derive monthly industry-specific interest and likability as a time series data type, which is a new industry evaluation indicator based on unstructured data. In addition, we propose a method to predict the industrial risk index, which is used as an important factor in industrial evaluation, based on derived industry-specific interest and likability time series data.

Advances in media technologies brought 3-D TV home and 4-D movies to your neighbour. We present a framework for 4-D broadcasting to bring 4-D entertainment home based on MPEG-V standard. A complete framework for 4-D entertainment from authoring of sensory effects to environment description and commanding rendering devices for the sensory effects can supported by MPEG-V and couple of other standards. Part 2 of MPEG-V provides tools for describing capabilities of the sensory devices and sensors, part 3 provides tools to describe sensory effects, and part 5 provides tools to actually interact with the sensory devices and sensors.

Previously a great attention has been paid to content-based multimedia retrieval that enables users to find and locate audio-visual material according to the intrinsic characteristics of the target. Query by humming (QBH) is also an application that makes retrieval according to major characteristics of music, that is, "melody". Our focus is on the realization of query-by-humming technology for easy-to-use application, which entails full automation of all the processes of the system, including melody information extraction from polyphonic audio. This paper presents developed algorithms for key steps of a QBH system including the melody feature extraction method from polyphonic audio and humming, their representation for matching, and matching method between the represented melody information from polyphonic audio and humming.

Almost every shot change detection algorithm detects abrupt transition (hard cut) without difficulty, but gradual transitions such as fades, dissolves, wipes are left as hard-to-detect problems. Dissolve effect, among the various gradual transition effects, is one of the most frequently used shot transition methods with special semantic meaning such as scene transition. Information of the shot change type can also be the basis for the shot clustering algorithms. In this paper, we present a fast and effective dissolve-transition detection algorithm based on the spatio-temporal distribution of the macro block pipes in MPEG-compressed video. In the proposed algorithm, the ratio of forward macro blocks in the B-type frames and the spatial distribution of forward/backward macro blocks are utilized for detecting dissolve transition. After finding such sequence of frames, we apply 2 heuristic rules: (1) The global color distributions of the frames at which dissolve starts and terminates are very different, and (2) The duration of dissolve transition is typically more than 0.3 second.