Rakuten Group, Inc.

Matsushita Electric Industrial Co., Ltd.

Ko Kikuchi

Mitsuo Makimoto

Sadahiko Yamashita

Haruyoshi Endo

A generator () generates a viewer video in which an object is synthesized at a predetermined synthesis position in a live video captured by an imager (). Additionally, the generator () generates a streamer video that is a video obtained by synthesizing the object in a mirror image video in which the live video is inverted, and in which the object is synthesized at a mirror image position relative to the synthesis position. A streaming unit () sends the viewer video generated by the generator () to a streaming server for streaming to the viewer terminal. A display) displays the streamer video generated by the generator ().

Streamer terminal, display method, and non-transitory computer readable medium for displaying video during live streaming

A distribution management system () plays, on a screen, a video (A) in which a plurality of topics (a, a, . . . ) is sequentially covered, determines a topic (a) currently being covered on the screen on which the video (A) is being played, identifies a list (L) of recommendation videos (X) in which a topic related to the determined topic (a) is covered, and provides the identified list (L) via the screen.

Distribution management for providing recommendation video in accordance with topic being viewed

The generator () generates a group (for example, a group chat) for viewers of a video introducing a product to interact with each other on social media (for example, a messenger service). The detector () detects a viewer who has viewed the video introducing the product and purchased the product. The registrar () then registers the viewer detected by the detector () to the group generated by the generator ().

Generation of group for viewers of video to communicate with each other

A collector () collects behaviors of a plurality of viewers viewing a live streaming video from a plurality of viewer terminals () respectively used by the plurality of viewers. The allocator () allocates, to each viewer of the plurality of viewers, another viewer specified as a referent from among the plurality of viewers based on the collected behaviors. The generator () generates display control information for each viewer based on a behavior of the viewer and a behavior of the referent allocated to the viewer among the collected behaviors. The transmitter () transmits the generated display control information to a viewer terminal () used by each viewer.

Video distribution in which display is controlled based on behavior of viewer

A distribution controller () distributes a live video provided from a streamer to a viewer terminal having a comment area, the comment area allowing a predetermined number of comments to be displayed. A receiver () receives a comment posted from the viewer terminal during distribution of the live video. A sender () sends a comment that the receiver () received to the viewer terminal to cause the comment to be added to the comment area in a last-in-first-out manner. An identifier () identifies a comment to which the streamer reacted in the live video being distributed by, every time the streamer speaks, comparing a speech text, the speech text being acquired by subjecting speech of the streamer to speech recognition, with respective comments currently being displayed in the comment area. A display controller () controls display of the identified comment.

Video distribution in which display of comments posted during distribution is controlled

A distribution controller () distributes a live video provided from a streamer to a plurality of terminals each of which has a comment area, the comment area allowing a predetermined number of comments to be displayed. A receiver () receives a comment posted from, among the plurality of terminals, a viewer terminal used by a viewer during distribution of the live video. A sender () sends a comment that the receiver () received to the plurality of terminals to cause the comment to be added to the comment areas in a last-in-first-out manner. An identifier () identifies, among respective comments currently being displayed in the comment area, a comment to which the streamer reacted. A setter () sets a chapter to divide the live video, based on a relative time in the live video corresponding to a time at which the identified comment is received.

Video distribution in which chapter is set based on reaction of streamer during distribution

A streamer terminal () makes (), upon receiving text, the text appear in a screen, determines () whether a streamer pays attention to the text, until the text exits from the screen, and executes () a predetermined attention process on text-of-attention that the streamer is determined to pay attention to. Each terminal () of viewer terminals classified into any one of groups transmits () text accepted from a viewer using each terminal to the streamer terminal () and another viewing terminals () belonging to an identical group, receives () text transmitted from another viewer terminal belonging to the identical group, and makes () the text appear in a screen. The streamer terminal () transmits () the text-of-attention to viewer terminals () classified in another group, and each terminal of the viewer terminals (), upon receiving the text-of-attention, makes the text-of-attention appear in the screen.

Live stream viewable while sending and receiving text in group and among groups

In an information processing device to streaming-distribute a video, based on a distance image in which a streamer is captured, an acquirer () acquires a depth distribution in the distance image. A determiner () determines a threshold value, based on the acquired depth distribution. An specifier () specifies, in the distance image, a distant region where depth is greater than the threshold value and a near region where depth is less than the threshold value. A provider () provides, in conjunction with the video, to a viewer terminal () to which the video is to be streaming-distributed, control information to set the distant region as a background, set the near region as a foreground, superimpose provided information over the background, and superimpose the foreground over the background over which the provided information is superimposed.

Video distribution in which provided information is superimposed on video

A band-stop filter for VHF-UHF band comprises at least three series resonance circuits and at least two transmission lines each connected between the series resonance circuits. One of the transmission lines, which is connected to the input terminal of the band-stop filter, has an electrical length which is shorter or longer than the quarter wavelength of the center frequency of the stop band by more than 20 but less than 50 percent. When a band-stop filter comprises four transmission lines, one of the transmission lines, which is connected to the output terminal of the band-stop filter, has the same length as the transmission line connected to the input terminal. With this arrangement, the band-stop filter exhibits a sharp attenuation characteristic in a frequency range below or above the center frequency. Other transmission lines, which are not directly connected to either the input terminal or the output terminal, may have a length which is shorter or longer than the quarter wavelength by 5 to 20 percent so as to provide a sharper attenuation characteristic.

Band-stop filter for VHF-UHF band

In a coaxial filter arrangement having a plurality of coaxial resonators each having a bottomed metallic cylinder interposed between a dielectric plate and one end of a center conductor of each resonator, the peripheral wall portion of the metallic cylinder is received in the bore of an outer conductor having an open end and a closed end connected to the other end of the center conductor. In one embodiment, the bottom of the metallic cylinder has a convex portion at its center so that the bottom center projects outwardly. In another embodiment, an annular metallic member is interposed between the dielectric plate and the outer surface of the flat bottom of the metallic cylinder. Since the distance between bottoms of metallic cylinders of adjacent resonators is made large with this arrangement, influence due to leakage flux of electromagnetic field is small, while stray capacitance is negligibly small, thereby providing a filter having high performance and low loss.

Growing community of inventors

Tokyo, Japan

Ko Kikuchi