English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Multimodal video-audio-text understanding and generation can benefit fromdatasets that are narrow but rich. The narrowness allows bite-sized challengesthat the research community can make progress on. The richness ensures we aremaking progress along the core challenges. To this end, we present alarge-scale video-audio-text dataset MUGEN, collected using the open-sourcedplatform game CoinRun [11]. We made substantial modifications to make the gamericher by introducing audio and enabling new interactions. We trained RL agentswith different objectives to navigate the game and interact with 13 objects andcharacters. This allows us to automatically extract a large collection ofdiverse videos and associated audio. We sample 375K video clips (3.2s each) andcollect text descriptions from human annotators. Each video has additionalannotations that are extracted automatically from the game engine, such asaccurate semantic maps for each frame and templated textual descriptions.Altogether, MUGEN can help progress research in many tasks in multimodalunderstanding and generation. We benchmark representative approaches on tasksinvolving video-audio-text retrieval and generation. Our dataset and code arereleased at: https://mugen-org.github.io/.

MUGEN: A Playground for Video-Audio-Text Multimodal Understanding and GENeration