Naver Corporation

About: Naver Corporation is a company organization based out in Seongnam-si, South Korea. It is known for research contribution in the topics: Terminal (electronics) & Computer science. The organization has 4038 authors who have published 4294 publications receiving 35045 citations. The organization is also known as: NAVER Corporation & NAVER.

System and method for providing information interactively by instant messaging application

Abstract: An information providing method and system that may provide information while conversing with a user through an instant messaging application is disclosed. The information providing system enables a user terminal to add, as a friend, a software bot which may be a virtual friend configured in an artificial software program through the instant messaging application executed in a user terminal. When the user inputs and transmits a message to the software bot through a conversation session, the information providing system may retrieve an appropriate response corresponding to the message from a database by inputting the message as a query, and may transmit the response to the user terminal, thereby providing the response to the user through the conversation session.

System and method for providing avatar with variable appearance

Abstract: The present invention relates a system and method for providing an avatar with variable appearance. When a user is connected through a network, information on the avatar provided to the user is collected. The avatar has at least two exposed units. The user's location on the network is determined, at least one exposed unit is selected from among the plurality of exposed units according to the determined location, and an avatar for displaying the selected exposed unit is generated to the user. The exposed units configuring the avatar are respectively modified according to the user's selection. Therefore, the avatar can be easily displayed on any place on the Internet by controlling the exposed units of the avatar depending on the user's location on the network. Further, the user can combine and modify the avatar in various ways by respectively controlling the exposed units.

Robust Image Retrieval-based Visual Localization using Kapture

Abstract: In this paper, we present a versatile method for visual localization. It is based on robust image retrieval for coarse camera pose estimation and robust local features for accurate pose refinement. Our method is top ranked on various public datasets showing its ability of generalization and its great variety of applications. To facilitate experiments, we introduce kapture, a flexible data format and processing pipeline for structure from motion and visual localization that is released open source. We furthermore provide all datasets used in this paper in the kapture format to facilitate research and data processing. Code and datasets can be found at this https URL, more information, updates, and news can be found at this https URL.

Combination of Multiple Global Descriptors for Image Retrieval

Abstract: Recent studies in image retrieval task have shown that ensembling different models and combining multiple global descriptors lead to performance improvement. However, training different models for the ensemble is not only difficult but also inefficient with respect to time and memory. In this paper, we propose a novel framework that exploits multiple global descriptors to get an ensemble effect while it can be trained in an end-to-end manner. The proposed framework is flexible and expandable by the global descriptor, CNN backbone, loss, and dataset. Moreover, we investigate the effectiveness of combining multiple global descriptors with quantitative and qualitative analysis. Our extensive experiments show that the combined descriptor outperforms a single global descriptor, as it can utilize different types of feature properties. In the benchmark evaluation, the proposed framework achieves the state-of-the-art performance on the CARS196, CUB200-2011, In-shop Clothes, and Stanford Online Products on image retrieval tasks. Our model implementations and pretrained models are publicly available.

System and method for sharing search result using messenger

Abstract: Disclose is a method and system for sharing search result using messenger. A messenger client generates search request information and receives web page data corresponding to the search request information. A search server receives the search request information from the messenger client and performs a search for one or more keywords included in the search request information to transmit search result page data to the messenger client. When the search result page is transmitted from the search server, a layer coupled to a conversation window is formed to display the search result page and allow users to share the information on web page corresponding to the search results when requested.