This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅ๐ฅMagicVideo: Text-To-Video #AI๐ฅ๐ฅ
๐MagicVideo: novel efficient text-to-video AI based on latent diffusion models
๐Review https://bit.ly/3UX5YCd
๐Paper arxiv.org/pdf/2211.11018.pdf
๐Project magicvideo.github.io/
๐MagicVideo: novel efficient text-to-video AI based on latent diffusion models
๐Review https://bit.ly/3UX5YCd
๐Paper arxiv.org/pdf/2211.11018.pdf
๐Project magicvideo.github.io/
๐ฑ8๐2๐ฉ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐งโโ๏ธ Generative Neural Texture for Avatars ๐งโโ๏ธ
๐Novel #3D GAN for (the most, ever) detailed facial avatars from unstructured 2D images
๐Review https://bit.ly/3EvF0dS
๐Project mrtornado24.github.io/Next3D/
๐Paper arxiv.org/pdf/2211.11208.pdf
๐Code github.com/MrTornado24/Next3D
๐Novel #3D GAN for (the most, ever) detailed facial avatars from unstructured 2D images
๐Review https://bit.ly/3EvF0dS
๐Project mrtornado24.github.io/Next3D/
๐Paper arxiv.org/pdf/2211.11208.pdf
๐Code github.com/MrTornado24/Next3D
๐ฅ6๐ฑ4๐2
๐ฅ๐ฅ "There can be only one" ๐ฅ๐ฅ
๐#Catfighting from the #AI community ๐
๐Review https://bit.ly/3Ax5I4n
๐#Catfighting from the #AI community ๐
๐Review https://bit.ly/3Ax5I4n
๐คฃ7๐5๐พ2
This media is not supported in your browser
VIEW IN TELEGRAM
๐๏ธ Neural Semantic Image Synthesis ๐๏ธ
๐Adobe unveils SceneComposer: text to 2D semantic canvas with precise shapes
๐Review https://bit.ly/3EtDyIL
๐Project zengyu.me/scenec
๐Paper arxiv.org/pdf/2211.11742.pdf
๐Code github.com/zengxianyu/scenec
๐Adobe unveils SceneComposer: text to 2D semantic canvas with precise shapes
๐Review https://bit.ly/3EtDyIL
๐Project zengyu.me/scenec
๐Paper arxiv.org/pdf/2211.11742.pdf
๐Code github.com/zengxianyu/scenec
๐คฏ10โค4โก1๐1
This media is not supported in your browser
VIEW IN TELEGRAM
๐น Perceptual NeRF-Inpainting ๐น
๐A novel complete process for #3D scene manipulation powered by NeRF
๐Review https://bit.ly/3Vn0EaT
๐Paper arxiv.org/pdf/2211.12254.pdf
๐Project spinnerf3d.github.io
๐A novel complete process for #3D scene manipulation powered by NeRF
๐Review https://bit.ly/3Vn0EaT
๐Paper arxiv.org/pdf/2211.12254.pdf
๐Project spinnerf3d.github.io
๐5๐ฅ3๐พ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅ Stable Diffusion 2.0 is out! ๐ฅ
๐The open-source of #StableDiffusion V2 just released. Magic ๐คฏ
๐Review https://bit.ly/3XwGInO
๐Code github.com/Stability-AI/stablediffusion
๐Blog stability.ai/blog/stable-diffusion-v2-release
๐The open-source of #StableDiffusion V2 just released. Magic ๐คฏ
๐Review https://bit.ly/3XwGInO
๐Code github.com/Stability-AI/stablediffusion
๐Blog stability.ai/blog/stable-diffusion-v2-release
๐คฏ17๐8๐พ4โก3๐ฑ3๐1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅMineDojo: Agents at Internet-Scale๐ฅ
๐The #neurips2022 Outstanding Paper Award winner is out. By #Nvidia et al.
๐Review https://bit.ly/3F0RPhN
๐Project https://minedojo.org/
๐Paper arxiv.org/pdf/2206.08853.pdf
๐Code github.com/MineDojo/MineDojo
๐The #neurips2022 Outstanding Paper Award winner is out. By #Nvidia et al.
๐Review https://bit.ly/3F0RPhN
๐Project https://minedojo.org/
๐Paper arxiv.org/pdf/2206.08853.pdf
๐Code github.com/MineDojo/MineDojo
๐ฅ10๐คฉ1
AI with Papers - Artificial Intelligence & Deep Learning
๐ท Pix2Seq: object detection by #Google ๐ท ๐A novel framework to perform object detection as a language modeling task ๐๐ข๐ ๐ก๐ฅ๐ข๐ ๐ก๐ญ๐ฌ: โ
Obj. detection as a lang-modeling task โ
BBs/labels -> seq. of discrete token โ
Encoder-decoder (one token at a time) โ
Code underโฆ
This media is not supported in your browser
VIEW IN TELEGRAM
โก4๐ฅฐ2๐1๐ฑ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ซ Tracking in Motion-Blurred Clips ๐ซ
๐Robust line segment detector in motion-blurred clips -> #SLAM / #3D.
๐Review https://bit.ly/3F2wHrb
๐Paper arxiv.org/pdf/2211.07365.pdf
๐Project levenberg.github.io/FE-LSD/
๐Code github.com/lh9171338/FE-LSD
๐Robust line segment detector in motion-blurred clips -> #SLAM / #3D.
๐Review https://bit.ly/3F2wHrb
๐Paper arxiv.org/pdf/2211.07365.pdf
๐Project levenberg.github.io/FE-LSD/
๐Code github.com/lh9171338/FE-LSD
๐คฏ5๐2๐ฅ2โก1
This media is not supported in your browser
VIEW IN TELEGRAM
๐EDGE: Diffusive Dancers Generator๐
๐New SOTA in editable human-dancer generation according to the input music
๐Review https://bit.ly/3u2egfY
๐Paper arxiv.org/pdf/2211.10658.pdf
๐Project edge-dance.github.io/
๐New SOTA in editable human-dancer generation according to the input music
๐Review https://bit.ly/3u2egfY
๐Paper arxiv.org/pdf/2211.10658.pdf
๐Project edge-dance.github.io/
โค8๐คฉ2๐คฏ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅฌ Diffusive Sketch-Guided Text-to-Image ๐ฅฌ
๐#Google unveils a universal approach for T2I (pre-trained) diffusion: free-hand, saliency-guided, etc.
๐Review https://bit.ly/3XFVMj2
๐Project sketch-guided-diffusion.github.io/
๐Paper sketch-guided-diffusion.github.io/files/sketch-guided-preprint.pdf
๐#Google unveils a universal approach for T2I (pre-trained) diffusion: free-hand, saliency-guided, etc.
๐Review https://bit.ly/3XFVMj2
๐Project sketch-guided-diffusion.github.io/
๐Paper sketch-guided-diffusion.github.io/files/sketch-guided-preprint.pdf
๐คฏ4โก1โค1๐1๐ฅ1๐ฅฐ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ SpaText: Spatio-Textual Generation ๐
๐#META AI unveils a novel T2I using open-vocabulary scene control
๐Review https://bit.ly/3OKcdXz
๐Project omriavrahami.com/spatext/
๐Paper arxiv.org/pdf/2211.14305.pdf
๐Code (coming)
๐#META AI unveils a novel T2I using open-vocabulary scene control
๐Review https://bit.ly/3OKcdXz
๐Project omriavrahami.com/spatext/
๐Paper arxiv.org/pdf/2211.14305.pdf
๐Code (coming)
๐คฏ7๐ฑ2โค1๐ฅ1๐คฉ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐กDiffusive Generation/Inpainting/Reconstruction๐ก
๐The first 3D-aware diffusion model purely trained from 2D images only
๐Review https://bit.ly/3XyHyjY
๐Paper arxiv.org/pdf/2211.09869.pdf
๐Code github.com/Anciukevicius/RenderDiffusion
๐The first 3D-aware diffusion model purely trained from 2D images only
๐Review https://bit.ly/3XyHyjY
๐Paper arxiv.org/pdf/2211.09869.pdf
๐Code github.com/Anciukevicius/RenderDiffusion
โค5๐2๐ฅ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ 3DDesigner: Text-guided Diffusion ๐
๐Novel method to generate #3D fine-grained objects via textual diffusion
๐Review https://bit.ly/3Fbmyc1
๐Project 3ddesigner-diffusion.github.io/
๐Paper arxiv.org/pdf/2211.14108.pdf
๐Novel method to generate #3D fine-grained objects via textual diffusion
๐Review https://bit.ly/3Fbmyc1
๐Project 3ddesigner-diffusion.github.io/
๐Paper arxiv.org/pdf/2211.14108.pdf
๐ฅ8๐2๐คฏ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐งจFast-SNARF: 150ร faster NeRF๐งจ
๐Voxel-based, pre-computed skinning, #cuda kernels -> avatar 150x faster!
๐Review https://bit.ly/3OMvEPo
๐Code github.com/xuchen-ethz/fast-snarf
๐Paper dataset.ait.ethz.ch/downloads/fast-snarf/paper.pdf
๐Voxel-based, pre-computed skinning, #cuda kernels -> avatar 150x faster!
๐Review https://bit.ly/3OMvEPo
๐Code github.com/xuchen-ethz/fast-snarf
๐Paper dataset.ait.ethz.ch/downloads/fast-snarf/paper.pdf
๐ฅ7๐คฉ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅ#AIwithPapers: we are 5,200+!๐ฅ
๐ Me (and you, all), these days (audio ON)๐คฃ
๐ Share -> https://bit.ly/3VBGN7Y
๐ Invite friends -> https://t.me/AI_DeepLearning
๐ Me (and you, all), these days (audio ON)๐คฃ
๐ Share -> https://bit.ly/3VBGN7Y
๐ Invite friends -> https://t.me/AI_DeepLearning
๐ฅฐ7๐คฃ5๐คฉ2๐ฑ1
๐ฅ๐ฅ Distilled #StableDiffusion paper! ๐ฅ๐ฅ
๐The progress with #StableDiffusion at #neurips2022 conference. Crazy times!
๐Review https://bit.ly/3ESo5Cn
๐Paper arxiv.org/pdf/2210.03142.pdf
๐The progress with #StableDiffusion at #neurips2022 conference. Crazy times!
๐Review https://bit.ly/3ESo5Cn
๐Paper arxiv.org/pdf/2210.03142.pdf
๐18๐คฏ3๐ฅ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ SJC: from 2D -> #3D generation ๐
๐ Novel method to convert pretrained 2D diffusion into 3D generative mode
๐Review https://bit.ly/3UppHJx
๐Paper arxiv.org/pdf/2212.00774.pdf
๐Code github.com/pals-ttic/sjc
๐Project pals.ttic.edu/p/score-jacobian-chaining
๐ Novel method to convert pretrained 2D diffusion into 3D generative mode
๐Review https://bit.ly/3UppHJx
๐Paper arxiv.org/pdf/2212.00774.pdf
๐Code github.com/pals-ttic/sjc
๐Project pals.ttic.edu/p/score-jacobian-chaining
โค2โก2๐1๐ฅ1๐คฉ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐คช Just another ORDINARY week in #AI community ๐คช
๐ Let's talk -> https://bit.ly/3VsEfth
๐ Invite friends -> https://t.me/AI_DeepLearning
๐ Let's talk -> https://bit.ly/3VsEfth
๐ Invite friends -> https://t.me/AI_DeepLearning
๐คฃ10๐1๐ฅ1