This media is not supported in your browser
VIEW IN TELEGRAM
๐ฏModeling the Human Pose Manifolds๐ฏ
๐#Meta Pose-NDF: continuous model for plausible human poses based on neural distance fields
๐Review https://bit.ly/3f6X59o
๐Paper arxiv.org/pdf/2207.13807.pdf
๐Project virtualhumans.mpi-inf.mpg.de/posendf
๐Code github.com/garvita-tiwari/PoseNDF
๐#Meta Pose-NDF: continuous model for plausible human poses based on neural distance fields
๐Review https://bit.ly/3f6X59o
๐Paper arxiv.org/pdf/2207.13807.pdf
๐Project virtualhumans.mpi-inf.mpg.de/posendf
๐Code github.com/garvita-tiwari/PoseNDF
โค6๐คฏ4
This media is not supported in your browser
VIEW IN TELEGRAM
๐ Gesture Recognition in 80's ๐
๐The #Casio AT-550 was offering the edge gesture recognition in 1984!
๐Review https://bit.ly/3fcPia6
๐Clip: youtube.com/watch?v=piFaJmYpQfQ
๐The #Casio AT-550 was offering the edge gesture recognition in 1984!
๐Review https://bit.ly/3fcPia6
๐Clip: youtube.com/watch?v=piFaJmYpQfQ
๐คฏ15๐ฅฐ6๐ฅ2
This media is not supported in your browser
VIEW IN TELEGRAM
๐ธMUSIKA! Neural โ-audio generation๐ธ
๐Novel neural music generation system on single consumer GPU!
๐Listen: https://bit.ly/3W80p4U
๐Paper arxiv.org/pdf/2208.08706.pdf
๐Code github.com/marcoppasini/musika
๐Novel neural music generation system on single consumer GPU!
๐Listen: https://bit.ly/3W80p4U
๐Paper arxiv.org/pdf/2208.08706.pdf
๐Code github.com/marcoppasini/musika
๐คฏ8
This media is not supported in your browser
VIEW IN TELEGRAM
๐จ๏ธMulti NeRF-supervised Depth-Pose๐จ๏ธ
๐NeRF-supervised disentanglement of depth/camera pose from large-scale clips
๐Review https://bit.ly/3gS4Z7e
๐Paper arxiv.org/pdf/2210.07181.pdf
๐Project oasisyang.github.io/self-mpinerf
๐NeRF-supervised disentanglement of depth/camera pose from large-scale clips
๐Review https://bit.ly/3gS4Z7e
๐Paper arxiv.org/pdf/2210.07181.pdf
๐Project oasisyang.github.io/self-mpinerf
๐ฑ4
This media is not supported in your browser
VIEW IN TELEGRAM
๐ง PlanT: Driving Plan w/ Transformer ๐ง
๐PlanT: planning via transformer. Imitation learning with compact object-level representation
๐Review https://bit.ly/3Ni2MxN
๐Project www.katrinrenz.de/plant/
๐Paper arxiv.org/pdf/2210.14222.pdf
๐Code github.com/autonomousvision/plant
๐PlanT: planning via transformer. Imitation learning with compact object-level representation
๐Review https://bit.ly/3Ni2MxN
๐Project www.katrinrenz.de/plant/
๐Paper arxiv.org/pdf/2210.14222.pdf
๐Code github.com/autonomousvision/plant
๐ฅ6๐2
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅNetwork -> C++ Converter. Pure Fire!๐ฅ
๐Fully open source #Python framework to convert Neural Nets into C++. Lightning-fast inference๐ฅ
๐Review https://bit.ly/3gSyqWI
๐Code github.com/facebookincubator/AITemplate
๐Fully open source #Python framework to convert Neural Nets into C++. Lightning-fast inference๐ฅ
๐Review https://bit.ly/3gSyqWI
๐Code github.com/facebookincubator/AITemplate
โค21๐คฏ12๐3
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅTAVA for photo-realistic Avatars๐ฅ
๐#META unveils a novel approach for Template-free Animatable Volumetric Avatars. #Metaverse ready.
๐Review https://bit.ly/3DPDEvm
๐Paper arxiv.org/pdf/2206.08929.pdf
๐Project www.liruilong.cn/projects/tava/
๐Code github.com/facebookresearch/tava
๐#META unveils a novel approach for Template-free Animatable Volumetric Avatars. #Metaverse ready.
๐Review https://bit.ly/3DPDEvm
๐Paper arxiv.org/pdf/2206.08929.pdf
๐Project www.liruilong.cn/projects/tava/
๐Code github.com/facebookresearch/tava
๐ฅฐ8๐1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ DANNCE: animal keypoints tracking ๐
๐DANNCE: anatomical #3D tracking across species and behaviors
๐Review https://bit.ly/3Nrx0P3
๐Code https://github.com/spoonsso/dannce
๐DANNCE: anatomical #3D tracking across species and behaviors
๐Review https://bit.ly/3Nrx0P3
๐Code https://github.com/spoonsso/dannce
๐ฅ9๐6
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅปMulti-Garment Virtual Try On (VTON)๐ฅป
๐URJC unveils ULNeFs, a novel approach to efficiently solve mix-and-match VTO for multiple garment layers. SUIT UP!
๐Review https://bit.ly/3sVVAOC
๐Project mslab.es/projects/ULNeF
๐URJC unveils ULNeFs, a novel approach to efficiently solve mix-and-match VTO for multiple garment layers. SUIT UP!
๐Review https://bit.ly/3sVVAOC
๐Project mslab.es/projects/ULNeF
๐7โค4
โ ๏ธ #AI art is making creatives obsolete โ ๏ธ
1. Scientist: "AI, let's make creatives obsolete"
2. AI: "Ok, let's do it! We have a deal"
๐ The AI ๐
๐More https://bit.ly/3DBWXa2
1. Scientist: "AI, let's make creatives obsolete"
2. AI: "Ok, let's do it! We have a deal"
๐ The AI ๐
๐More https://bit.ly/3DBWXa2
๐คฃ17
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฉณ STC: Transformers MOT ๐ฉณ
๐A novel Multi-Object Tracking based on Transformers with Dense Representations
๐Review https://bit.ly/3WzQsNB
๐Paper arxiv.org/pdf/2210.13570.pdf
๐Code github.com/amitgalor18/STC_Tracker
๐A novel Multi-Object Tracking based on Transformers with Dense Representations
๐Review https://bit.ly/3WzQsNB
๐Paper arxiv.org/pdf/2210.13570.pdf
๐Code github.com/amitgalor18/STC_Tracker
๐คฏ16๐2
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅ#AIwithPapers: we are 4,700!๐ฅ
๐ Me, literally, every day ๐
๐ Share -> https://bit.ly/3fA5geu
๐ Invite friends -> https://t.me/AI_DeepLearning
๐ Me, literally, every day ๐
๐ Share -> https://bit.ly/3fA5geu
๐ Invite friends -> https://t.me/AI_DeepLearning
๐คฃ29๐8โค3๐ฅ2โก1๐คฉ1๐พ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐First Depth/anti-FacRec Cam ๐
๐First working prototype that enables passive depth estimation while inhibiting face identification
๐Code (coming)
๐Project https://zaidtas.github.io/privacymask
๐Paper zaidtas.github.io/privacymask/pdf/privacy_preserving_depth_estimation.pdf
๐First working prototype that enables passive depth estimation while inhibiting face identification
๐Code (coming)
๐Project https://zaidtas.github.io/privacymask
๐Paper zaidtas.github.io/privacymask/pdf/privacy_preserving_depth_estimation.pdf
๐ฅ5๐2๐คฉ2
This media is not supported in your browser
VIEW IN TELEGRAM
๐ Synthetic Expression-Wrinkles ๐
๐#Microsoft unveils a novel approach that produces realistic wrinkles across humans
๐Review https://bit.ly/3zWZLOd
๐Paper arxiv.org/pdf/2210.03529.pdf
๐Project microsoft.github.io/DynamicWrinkles
๐#Microsoft unveils a novel approach that produces realistic wrinkles across humans
๐Review https://bit.ly/3zWZLOd
๐Paper arxiv.org/pdf/2210.03529.pdf
๐Project microsoft.github.io/DynamicWrinkles
๐ฅ7๐คฏ4๐2๐ฑ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐FactorMatte Video re-Composition๐
๐FactorMatte: alternative formulation of video matting problem for re-composition
๐Review https://bit.ly/3NX6hdg
๐Paper arxiv.org/pdf/2211.02145.pdf
๐Project https://factormatte.github.io/
๐Code github.com/jacklyngu/FactorMatte
๐FactorMatte: alternative formulation of video matting problem for re-composition
๐Review https://bit.ly/3NX6hdg
๐Paper arxiv.org/pdf/2211.02145.pdf
๐Project https://factormatte.github.io/
๐Code github.com/jacklyngu/FactorMatte
๐คฏ9๐ฅ7
This media is not supported in your browser
VIEW IN TELEGRAM
๐ Unifying Flow, Stereo & Depth ๐
๐A unified neural model for motion and #3D perception tasks
๐Review https://bit.ly/3AbdnFk
๐Paper arxiv.org/pdf/2211.05783.pdf
๐Project haofeixu.github.io/unimatch/
๐Code github.com/autonomousvision/unimatch
๐A unified neural model for motion and #3D perception tasks
๐Review https://bit.ly/3AbdnFk
๐Paper arxiv.org/pdf/2211.05783.pdf
๐Project haofeixu.github.io/unimatch/
๐Code github.com/autonomousvision/unimatch
๐11โก2
This media is not supported in your browser
VIEW IN TELEGRAM
๐ซ Real-Time Auto Neural Mapping ๐ซ
๐Robot can densely annotate a scene via fully-autonomous experimental interactions
๐Review https://bit.ly/3GfwQbY
๐Paper arxiv.org/pdf/2210.17325.pdf
๐Project ihaughton.github.io/RobE
๐Robot can densely annotate a scene via fully-autonomous experimental interactions
๐Review https://bit.ly/3GfwQbY
๐Paper arxiv.org/pdf/2210.17325.pdf
๐Project ihaughton.github.io/RobE
๐ฅ8โก2
This media is not supported in your browser
VIEW IN TELEGRAM
๐๏ธ InfiniteNature-Zero: โ Video ๐๏ธ
๐Open-source generative #AI for unbounded flythrough videos of natural scenes
๐Review https://bit.ly/3hJYELa
๐Paper arxiv.org/pdf/2207.11148.pdf
๐Project infinite-nature-zero.github.io
๐Code https://github.com/google-research/google-research/tree/master/infinite_nature_zero
๐Open-source generative #AI for unbounded flythrough videos of natural scenes
๐Review https://bit.ly/3hJYELa
๐Paper arxiv.org/pdf/2207.11148.pdf
๐Project infinite-nature-zero.github.io
๐Code https://github.com/google-research/google-research/tree/master/infinite_nature_zero
โค12๐4๐ฅ2โก1๐ฅฐ1๐1
This media is not supported in your browser
VIEW IN TELEGRAM
๐งถOpen-Source Neural Inpainting Suite๐งถ
๐Lama Cleaner: a free and open-source inpainting tool powered by SOTA #AI model
๐Review https://bit.ly/3Ex2aSq
๐Project github.com/Sanster/lama-cleaner
๐Lama Cleaner: a free and open-source inpainting tool powered by SOTA #AI model
๐Review https://bit.ly/3Ex2aSq
๐Project github.com/Sanster/lama-cleaner
๐คฏ12๐พ2๐ฅฐ1