🛢️🪵#Google is open-sourcing SAVi, novel approach to extend the Slot Attention mechanism to videos🛢️🪵

👉Object-centric representations are a promising...

🚀🚀#Microsoft open-sourced NÜWA, unified framework for neural visual world creation (better than DALL·E)🚀🚀

👉Unified multimodal pretrained model to… | 12 comments on LinkedIn

🪐🌴IceVision is the first agnostic Computer Vision framework with hundreds of pre-trained models🪐🌴

👉A curated collection with hundreds of HQ pre-trained...

🚜🚜ETH Zürich open-sourced PCAN, neural framework for online multiple object tracking & segmentation🚜🚜

👉MOT and segmentation requires detecting, tracking,… | 18 comments on LinkedIn

🍘🍘ICL unveils a smart labeler by Semantic-NeRF🍘🍘

👉Extending NeRF to jointly encode semantics with appearance/geometry and deal with efficient scene…

🥞🥞BlenderProc2: a procedural #Blender pipeline for photorealistic neural rendering🥞🥞

👉BlenderProc helps in generating real looking images for the...

🧁🧁The new SOTA in RVOS (semantic segmentation + captions) just open-sourced🧁🧁

👉RVOS task involves segmentation of a text-referred object instance in the… | 22 comments on LinkedIn

🥝🥝HyperStyle, a hypernetwork to modulate StyleGAN for Hi-Fi inversion, just open-sourced🥝🥝

👉Inversion of real images into StyleGAN’s latent space is an…

🧎🧍🤸CMU + #Nvidia unveil GLAMR: #3D global human mesh tracking in dynamic scenarios🧎🧍🤸

👉Recovering fine-grained 3D human meshes is essential for…

👽👽SAFA, a novel structure-aware neural generator, just open-sourced👽👽

👉GAN has made great progress on the face animation task. However, without knowing...

🔴🟢 Berkeley unveils PHALP, a novel 3D tracking by appearance, location & pose 🔴🟢

👉Novel approach for tracking people in monocular videos by predicting… | 10 comments on LinkedIn

🚶‍♂️🚶🚶‍♀️ Do pedestrians pay attention? 🚶‍♂️🚶🚶‍♀️

👉 Humans rely on eye contact for fast/efficient communication with nearby people. Autonomous...

🟦🟦 #META (Facebook AI) unlocks HM3D, the world’s largest indoor spaces dataset ever 🟦🟦

👉Facebook #AI and Matterport have collaborated on the release...

🍉🍉 Plenoxel: Radiance Fields 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 neural networks 🍉🍉

👉Berkeley unveils Plenoxel, a view-dependent sparse voxel model that can optimize ...

500 million+ members | Manage your professional identity. Build and engage with your professional network. Access knowledge, insights and opportunities.

 👗🥻👘 SenseTime open-sourced 𝐆𝐚𝐫𝐦𝐞𝐧𝐭𝟒𝐃, #AI for human-garment reconstruction👗🥻👘

👉Learning to reconstruct 3D garments is important for dressing...

😈👿 HeadNeRF: NeRF-based #AI for human faces 😈👿

👉Chinese researchers unveil a novel NeRF-based method to render Hi-Fi head images in real-time, and...

🍠🍠 ViT Deep features for Co-segmentation, Part Co-segmentation, and Semantic Correspondences 🍠🍠

👉“Deep Features” extracted from the activation of layers…

🥭🥭Real-ESRGAN: Super-Resolution with pure Synthetic Data🥭🥭

👉Single image super-resolution topic aims at reconstructing a high-resolution image from its… | 16 comments on LinkedIn

🎁🎁EgoBody: dataset for #3d human interactions🎁🎁

👉Understanding the social interactions from first-person view is crucial for robotics, #AR #VR and more…