This media is not supported in your browser
VIEW IN TELEGRAM
π₯MineDojo: Agents at Internet-Scaleπ₯
πThe #neurips2022 Outstanding Paper Award winner is out. By #Nvidia et al.
πReview https://bit.ly/3F0RPhN
πProject https://minedojo.org/
πPaper arxiv.org/pdf/2206.08853.pdf
πCode github.com/MineDojo/MineDojo
πThe #neurips2022 Outstanding Paper Award winner is out. By #Nvidia et al.
πReview https://bit.ly/3F0RPhN
πProject https://minedojo.org/
πPaper arxiv.org/pdf/2206.08853.pdf
πCode github.com/MineDojo/MineDojo
π₯10π€©1
AI with Papers - Artificial Intelligence & Deep Learning
π· Pix2Seq: object detection by #Google π· πA novel framework to perform object detection as a language modeling task ππ’π π‘π₯π’π π‘ππ¬: β
Obj. detection as a lang-modeling task β
BBs/labels -> seq. of discrete token β
Encoder-decoder (one token at a time) β
Code underβ¦
This media is not supported in your browser
VIEW IN TELEGRAM
β‘4π₯°2π1π±1
This media is not supported in your browser
VIEW IN TELEGRAM
π« Tracking in Motion-Blurred Clips π«
πRobust line segment detector in motion-blurred clips -> #SLAM / #3D.
πReview https://bit.ly/3F2wHrb
πPaper arxiv.org/pdf/2211.07365.pdf
πProject levenberg.github.io/FE-LSD/
πCode github.com/lh9171338/FE-LSD
πRobust line segment detector in motion-blurred clips -> #SLAM / #3D.
πReview https://bit.ly/3F2wHrb
πPaper arxiv.org/pdf/2211.07365.pdf
πProject levenberg.github.io/FE-LSD/
πCode github.com/lh9171338/FE-LSD
π€―5π2π₯2β‘1
This media is not supported in your browser
VIEW IN TELEGRAM
πEDGE: Diffusive Dancers Generatorπ
πNew SOTA in editable human-dancer generation according to the input music
πReview https://bit.ly/3u2egfY
πPaper arxiv.org/pdf/2211.10658.pdf
πProject edge-dance.github.io/
πNew SOTA in editable human-dancer generation according to the input music
πReview https://bit.ly/3u2egfY
πPaper arxiv.org/pdf/2211.10658.pdf
πProject edge-dance.github.io/
β€8π€©2π€―1
This media is not supported in your browser
VIEW IN TELEGRAM
π₯¬ Diffusive Sketch-Guided Text-to-Image π₯¬
π#Google unveils a universal approach for T2I (pre-trained) diffusion: free-hand, saliency-guided, etc.
πReview https://bit.ly/3XFVMj2
πProject sketch-guided-diffusion.github.io/
πPaper sketch-guided-diffusion.github.io/files/sketch-guided-preprint.pdf
π#Google unveils a universal approach for T2I (pre-trained) diffusion: free-hand, saliency-guided, etc.
πReview https://bit.ly/3XFVMj2
πProject sketch-guided-diffusion.github.io/
πPaper sketch-guided-diffusion.github.io/files/sketch-guided-preprint.pdf
π€―4β‘1β€1π1π₯1π₯°1
This media is not supported in your browser
VIEW IN TELEGRAM
π SpaText: Spatio-Textual Generation π
π#META AI unveils a novel T2I using open-vocabulary scene control
πReview https://bit.ly/3OKcdXz
πProject omriavrahami.com/spatext/
πPaper arxiv.org/pdf/2211.14305.pdf
πCode (coming)
π#META AI unveils a novel T2I using open-vocabulary scene control
πReview https://bit.ly/3OKcdXz
πProject omriavrahami.com/spatext/
πPaper arxiv.org/pdf/2211.14305.pdf
πCode (coming)
π€―7π±2β€1π₯1π€©1
This media is not supported in your browser
VIEW IN TELEGRAM
π‘Diffusive Generation/Inpainting/Reconstructionπ‘
πThe first 3D-aware diffusion model purely trained from 2D images only
πReview https://bit.ly/3XyHyjY
πPaper arxiv.org/pdf/2211.09869.pdf
πCode github.com/Anciukevicius/RenderDiffusion
πThe first 3D-aware diffusion model purely trained from 2D images only
πReview https://bit.ly/3XyHyjY
πPaper arxiv.org/pdf/2211.09869.pdf
πCode github.com/Anciukevicius/RenderDiffusion
β€5π2π₯1
This media is not supported in your browser
VIEW IN TELEGRAM
π 3DDesigner: Text-guided Diffusion π
πNovel method to generate #3D fine-grained objects via textual diffusion
πReview https://bit.ly/3Fbmyc1
πProject 3ddesigner-diffusion.github.io/
πPaper arxiv.org/pdf/2211.14108.pdf
πNovel method to generate #3D fine-grained objects via textual diffusion
πReview https://bit.ly/3Fbmyc1
πProject 3ddesigner-diffusion.github.io/
πPaper arxiv.org/pdf/2211.14108.pdf
π₯8π2π€―1
This media is not supported in your browser
VIEW IN TELEGRAM
π§¨Fast-SNARF: 150Γ faster NeRFπ§¨
πVoxel-based, pre-computed skinning, #cuda kernels -> avatar 150x faster!
πReview https://bit.ly/3OMvEPo
πCode github.com/xuchen-ethz/fast-snarf
πPaper dataset.ait.ethz.ch/downloads/fast-snarf/paper.pdf
πVoxel-based, pre-computed skinning, #cuda kernels -> avatar 150x faster!
πReview https://bit.ly/3OMvEPo
πCode github.com/xuchen-ethz/fast-snarf
πPaper dataset.ait.ethz.ch/downloads/fast-snarf/paper.pdf
π₯7π€©1
This media is not supported in your browser
VIEW IN TELEGRAM
π₯#AIwithPapers: we are 5,200+!π₯
π Me (and you, all), these days (audio ON)π€£
π Share -> https://bit.ly/3VBGN7Y
π Invite friends -> https://t.me/AI_DeepLearning
π Me (and you, all), these days (audio ON)π€£
π Share -> https://bit.ly/3VBGN7Y
π Invite friends -> https://t.me/AI_DeepLearning
π₯°7π€£5π€©2π±1
π₯π₯ Distilled #StableDiffusion paper! π₯π₯
πThe progress with #StableDiffusion at #neurips2022 conference. Crazy times!
πReview https://bit.ly/3ESo5Cn
πPaper arxiv.org/pdf/2210.03142.pdf
πThe progress with #StableDiffusion at #neurips2022 conference. Crazy times!
πReview https://bit.ly/3ESo5Cn
πPaper arxiv.org/pdf/2210.03142.pdf
π18π€―3π₯1
This media is not supported in your browser
VIEW IN TELEGRAM
π SJC: from 2D -> #3D generation π
π Novel method to convert pretrained 2D diffusion into 3D generative mode
πReview https://bit.ly/3UppHJx
πPaper arxiv.org/pdf/2212.00774.pdf
πCode github.com/pals-ttic/sjc
πProject pals.ttic.edu/p/score-jacobian-chaining
π Novel method to convert pretrained 2D diffusion into 3D generative mode
πReview https://bit.ly/3UppHJx
πPaper arxiv.org/pdf/2212.00774.pdf
πCode github.com/pals-ttic/sjc
πProject pals.ttic.edu/p/score-jacobian-chaining
β€2β‘2π1π₯1π€©1
This media is not supported in your browser
VIEW IN TELEGRAM
π€ͺ Just another ORDINARY week in #AI community π€ͺ
π Let's talk -> https://bit.ly/3VsEfth
π Invite friends -> https://t.me/AI_DeepLearning
π Let's talk -> https://bit.ly/3VsEfth
π Invite friends -> https://t.me/AI_DeepLearning
π€£10π1π₯1
Should I write a book (no technicalities, for everyone) about the evolution of the modern Computer Vision / AI ?
Anonymous Poll
81%
Yes, please!π§‘
19%
No, c'mon. Comment -> https://bit.ly/3B6T6S9
β€9π₯3π₯°1
This media is not supported in your browser
VIEW IN TELEGRAM
π¨ Dr.3D: adapting 3D GANs to Arts π¨
πNovel 3D GAN domain adaptation method for drawings. From portraits to 3D!
πReview https://bit.ly/3He3Phv
πPaper arxiv.org/pdf/2211.16798.pdf
πProject jinwonjoon.github.io/dr3d/
πCode github.com/JinWonjoon/Dr.3D
πNovel 3D GAN domain adaptation method for drawings. From portraits to 3D!
πReview https://bit.ly/3He3Phv
πPaper arxiv.org/pdf/2211.16798.pdf
πProject jinwonjoon.github.io/dr3d/
πCode github.com/JinWonjoon/Dr.3D
π₯°4π₯2β‘1π1π€―1
This media is not supported in your browser
VIEW IN TELEGRAM
π« Tracking EVERYTHING in the wild π«
πDisentangling classification from tracking, Track Every Thing Accuracy (TETA) for SOTA
πReview https://bit.ly/3F2rfU7
πPaper arxiv.org/pdf/2207.12978.pdf
πProject www.vis.xyz/pub/tet/
πDisentangling classification from tracking, Track Every Thing Accuracy (TETA) for SOTA
πReview https://bit.ly/3F2rfU7
πPaper arxiv.org/pdf/2207.12978.pdf
πProject www.vis.xyz/pub/tet/
π€―4π1π₯1π€©1
This media is not supported in your browser
VIEW IN TELEGRAM
π½HD Faces via "Hacked" Smartphoneπ½
πA novel method for HD faces from (polarizing-foils) smartphones
πReview https://bit.ly/3H8yhJN
πPaper arxiv.org/pdf/2212.01160.pdf
πProject dazinovic.github.io/polface
πA novel method for HD faces from (polarizing-foils) smartphones
πReview https://bit.ly/3H8yhJN
πPaper arxiv.org/pdf/2212.01160.pdf
πProject dazinovic.github.io/polface
π₯4π€―1π±1
This media is not supported in your browser
VIEW IN TELEGRAM
π Clothes-Changing Re-Identification π
πThe new SOTA in wild Re-identification by mixing appearance & face features
πReview https://bit.ly/3HdTTUU
πPaper arxiv.org/pdf/2211.13807.pdf
πProject www.vision.huji.ac.il/reface
πCode github.com/bar371/ReFace
πThe new SOTA in wild Re-identification by mixing appearance & face features
πReview https://bit.ly/3HdTTUU
πPaper arxiv.org/pdf/2211.13807.pdf
πProject www.vision.huji.ac.il/reface
πCode github.com/bar371/ReFace
π5π€―3β‘2π±1
This media is not supported in your browser
VIEW IN TELEGRAM
π SOTA: One-Shot Animatable Avatars π
πELICIT: synthesis & avatar animation -> NeRF representation in one-shot
πReview https://bit.ly/3P3eZY6
πPaper arxiv.org/pdf/2212.02469.pdf
πProject https://elicit3d.github.io/
πCode github.com/huangyangyi/ELICIT
πELICIT: synthesis & avatar animation -> NeRF representation in one-shot
πReview https://bit.ly/3P3eZY6
πPaper arxiv.org/pdf/2212.02469.pdf
πProject https://elicit3d.github.io/
πCode github.com/huangyangyi/ELICIT
π₯4π2β€1π₯°1π±1π©1