Research

関連キーワード

エンタテインメントコンピューティング,拡張現実,O2O,ユーザインタフェース,HCI,eコマース,ソーシャルサービス,画像解析,オンライン文字認識,画像内文字検索,OCR,生体情報解析,ゲームテクノロジー,ショッピングエンタテインメント,コンピュータグラフィックス,眼球運動,バーチャルヒューマン,バーチャルファッション,ファッションコーディネート,3Dアニメーション,情報可視化

研究動画など

Rakuten Institute of Technology Launches Interactive Game for Rakuten Eagles Fans

Rakuten Institute of Technology launched the interactive game “DYNAMITE BOAT RACE Inning,” playable at the Rakuten Kobo Stadium Miyagi. The app has adopted the smartphone-related technology that was developed by Rakuten Institute of Technology, and can be played between innings on the giant LED screen during Tohoku Rakuten Golden Eagles home games. This is the same technology that has been used for “WallSHOP,” the digital shopping signage that is in use at Rakuten Crimson House.
Check out this video for an overview of Rakuten’s first try at providing a real-time game as well as the background of its development.

RIT’s New Digital Signage at Rakuten Crimson House

In the first floor lobby of Rakuten Crimson House, which opened on June 22, is a display of WallSHOP, the latest in digital signage.
Let’s take a look back at the history of digital signage at RIT so far.

WallSHOP meets RAKUMA: Displaying RAKUMA-listed items on Digital Signage

RAKUMA (Rakuten Flea Market application) items are displayed on touch-based WallSHOP at Rakuten Tower Entrance.
This is one of experiences REAL to App.

WallSHOP: Multiuser Interaction with Public Digital Signage using Private Mobile Devices

WallSHOP bridges the public-to-private experience in e-commerce using interconnectivity of digital signage and personal mobile devices. Multiple users can connect and interact with a wall display and pull public content to their mobile devices.

Texts search in Images Using Pattern Similarity of Recognized Characters

Many online shops tend to use images embedded with texts, which cannot be searched by conventional text search technologies.
We propose a novel text search system that scores images based on:
• Pattern similarity of recognized characters
• Visual saliency
• Term frequency
We applied our approach within an e-commerce shopping mall system which has large amount of images used in shops and product pages.

Enriching Event Information Display with Touch Interaction

We developed a digital signage system which shows rich information of Rakuten merchants and their products. By touching displayed items, you can get detailed information such as online review comments, which were extracted automatically by Natural Language Processing technology. This signage system was exhibited at Umaimono Taikai held at Hakata Hankyu from March 12 to March 18 2014, which was one of the biggest food exhibitions powered by Rakuten.

*PicSer – Picture Based Search for Rakuten Ichiba

We provided a chrome extension named *PicSer, picture-based search result interface for Rakuten Ichiba. It helps to find your favorite items only by their appearance. *PicSer has these functions by using Image Processing technology and Natural Language Processing.
1. Rich Image Window Shopping
2. Color Filtering
3. Similar Item Search
4. Image Grouping
5. Compare Items
6. Keywords Addition & Reduction by Mouse Gesture

AR-HITOKE: Visualizing Popularity of Brick and Mortar Shops to Support Purchase Decisions

We propose a shopping support system (AR-HITOKE) that visualizes the popularity of brick and mortar shops by aggregating online and offline information using augmented reality technology that can be understood intuitively.
In Umaimono-Taikai Nagoya 2013 which is a food exhibition and sale powered by Rakuten, we provided this AR-HITOKE to visualize shops’ popularity using POS data.
Shoppers were able to “see” which shop is popular by the amount of human icons.

Visualization of dimension measurement using a consumer grade tablet camera-audio sensor

This is a dimension measurement visualization technique using a consumer tablet device. A user is able to control the position and size of a virtual dimension image and confirm the specified dimension on the camera view. The system utilizes speaker and microphone sensors as distance detector to extract information needed for correct sizing of the visualized measurement.

KooSHO: Japanese Text Input Environment based on Aerial Hand Writing

We propose KooSHO: an environment for Japanese text input based on aerial hand gestures. The system provides an integrated experience of character input, Kana-Kanji conversion, and search result visualization.

AR Hitoke: Visualization of Online Purchase Activities by Augumented Reality

HITOKE: A Study of Queue Visualization of Internet Purchase Information

The HITOKE interface visualizes purchase context such as buyer attributes or product purchase frequency on web pages as a queue representation, and has undertaken to provide users with new evaluation information.

KiTeMiROOM: A Fashion-Coordination System for Mobile Devices

This“KiTeMiROOM” interface allows you to quickly browse coordinating tops and bottoms clothes, as well as shoes by simultaneously displaying them using carousel panels.

AteGau: Projector-Based Online Fashion Coordination System

Our AteGau system enables intuitive fashion coordination using both
clothes from an online store and clothes from the user’s own wardrobe.

Lovable Couch: Mitigating Distrustful Feelings for Couples by Visualizing Excitation

We propose Lovable Couch, an approach to support dating session by visually actuating a sofa with user’s excitement measures as a way to mitigate users’ anxiety. Lovable Couch is a first step of a short distance communication system at speed dating.

実画像を利用した仮想ファッションシステム

virtualfashion

仮想俳優のための視線アニメーション生成技術

virtualhuman

心拍数を利用したフィットネス支援ゲーム

fitnessgame