ACM SIGGRAPH’s Post

this demo never gets old for me... check out claude 3.5 sonnet and its new multi-modal capability especially with interpreting charts, graphs, or transcribing text from imperfect images.

🛡️ Sentinel: "Sentinel captures the essence of vigilance and protection, evoking the idea of a lone figure standing guard over something precious or important. The abstract nature of the piece leaves room for interpretation, inviting viewers to ponder what this sentinel might be safeguarding." Stay tuned for more . ✨ #Art #DigitalArt #Sentinel #Abstractsurrealism

What does it mean to crip art, science, and technology knowledge and praxis? 𝗪𝗛𝗘𝗡: 26 April 2024 @ 10:30AM–12:00PM PDT (UTC -7) 𝗪𝗛𝗘𝗥𝗘: Online via Zoom (Registration Required) 𝗥𝗘𝗚𝗜𝗦𝗧𝗘𝗥 𝗡𝗢𝗪: https://buff.ly/3vt52y1 Join us for an exciting event celebrating the launch of the Leonardo special issue, "Criptech and the Art of Access," as we explore innovative approaches to accessibility, reimagining what access can be when it transcends basic accommodation or regulatory compliance. Featuring presentations and a roundtable discussion moderated by special issue editors Lindsey D. Felt and Vanessa Chang, along with contributors Laura Forlano, Marina Tsaplina, Ysolde Stienon, Aminder Virdee, Erika Jean Lincoln, and Darrin Martin. ASL interpretation and captioning provided. This event will be recorded for archival purposes. Don't miss out – RSVP today! Image credit: Itziar Barrio,Artefact II, 2023, custom circuit board, concrete, epoxy, and Laura Forlano’s text, 34 x 34 x 7.5 cm (13.375 x 13.375 x 3 in). (Photo: EtienneFrossard.) #CripTech #Access #DisabilityInnovation #Metaverse #ArtSciTech #Publications

#ipsarc14 auspiced by #USMF_Moldova, coordinated by @T_Agarin under the slogan "Challenges of Plural Societies". The term 𝕮𝖍𝖆𝖑𝖑𝖊𝖓𝖌𝖊, through Old 🇫🇷 comes from the Latin ℭ𝔞𝔩𝔲𝔪𝔫𝔦𝔞𝔯𝔦, to calumniate, not just to accuse or object, but to 𝐚𝐜𝐜𝐮𝐬𝐞 𝐟𝐚𝐥𝐬𝐞𝐥𝐲 I'm taking part with a paper on "Iraq: Navigating Postcolonial Calamities and Structural Failures" at session 2B "Challenges for démocratisation in the post-soviet space ", and will be glad to chair and discuss at panels 3C and 3D

layers of atmosphere model 3d - science project for exhibition - diy - simple and easy steps #layersofatmosphere #atomspheremodel #3d #scienceexhibition #sciencefair #howtofunda https://lnkd.in/g8Z2VAM6

Are you heading to #Siggraph2024? So are we! Our very own Bobby Carlton is part of this years program as part of a panel that also includes experts from NVIDIA, Meta, and Amazon discussing how to expand 3D skills into AI, robotics, and digital twins solutions. See his session on Tuesday at 1pm MDT. #digitaltwins #ai #robotics

WIP002 Houdini & Octane Research on vertex maps and the development of mixed materials

The re-emergence of historical architecture in the next generation of technology and technological expression is a new beginning of introducing detail to the box. Like cave paintings, we are once again remembering why we drew them in the first place. Classical architecture, redefined, re-examined, re-explained, newly understood.

Have you ever tried to create depth from monocular video? If so, you’re probably familiar with #Midas, #Marigold, #DepthAnything, etc. While they can produce great results on still frames, they often suffer from temporal consistency issues when applied to image sequences. To solve this problem, #ChronoDepth leverages a pre-trained video diffusion model, Stable Video Diffusion. The approach involves a two-stage training process: first, optimizing spatial layers using single-frame depths, then freezing these layers and training temporal layers with video clips. This strategy enhances both spatial accuracy and temporal consistency. Additionally, a sliding window strategy for inference further improves consistency by inpainting frames based on previous predictions. Is it perfect? Definitely not! But the enhancement over existing models is quite remarkable, and I’m confident we will see even more consistent outputs in the near future. You can find more Information on the project page here: https://lnkd.in/eMfheMvg Run it yourself by cloning the repo: https://lnkd.in/eNG6fzAy Or check out the Gradio demo available on Hugging Face: https://lnkd.in/eaY4dMSx #chronodepth #svd #gradio #midas #marigold #depthanything #huggingface #depthfromvideo #stablediffusion 

Sasha Kojjio is an innovative artist known for his exploration of language, technology, and human perception through his audiovisual installations. Sasha Kojjio's artistic practice is based on three key elements – concept, aesthetics and program code. Sasha chose multimedia art as his main medium, because it allows him, through programming, to translate a visual message to create a fundamentally new experience for the viewer to interact with his art on a non-semantic level. "Semantic Failure," which premiered at NON SEMANTIC 2023, is a poignant reflection on the breakdown of communication in the digital era. The piece intertwines fragmented text with glitchy, abstract visuals and dissonant sounds, symbolising the distortion and loss of meaning in a world overwhelmed by information. https://lnkd.in/enfuM5zN #loveart #thattick #ImmersiveArt #artinstallation