This media is not supported in your browser
VIEW IN TELEGRAM
๐ชฌ META's Animated Drawings is out! ๐ชฌ
๐#META unveils an easy-to-use method for animating human-like figures drawn by children.
๐Review https://bit.ly/3mGeQQv
๐Paper arxiv.org/pdf/2303.12741.pdf
๐Project fairanimateddrawings.com
๐#META unveils an easy-to-use method for animating human-like figures drawn by children.
๐Review https://bit.ly/3mGeQQv
๐Paper arxiv.org/pdf/2303.12741.pdf
๐Project fairanimateddrawings.com
๐ฑ16๐ฅฐ5๐4๐2๐คฉ2โก1๐ฅ1๐พ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฆ 6D Non-Prehensile Manipulation ๐ฆ
๐#META (+CMU) unveils HACMan, novel 6D non-prehensile manipulation of objects
๐Review https://bit.ly/3NP1jl1
๐Paper arxiv.org/pdf/2305.03942.pdf
๐Project hacman-2023.github.io
๐#META (+CMU) unveils HACMan, novel 6D non-prehensile manipulation of objects
๐Review https://bit.ly/3NP1jl1
๐Paper arxiv.org/pdf/2305.03942.pdf
๐Project hacman-2023.github.io
๐6๐ฅ4๐คฏ3๐ฑ1
๐ฆ Llama-2: the Open-Source "ChatGPT" ๐ฆ
๐GenAI, #Meta unveils Llama-2: a collection of LLMs ranging in scale 7-70B params. Challenging with #chatgpt, but open.
๐Review https://t.ly/bLJgP
๐Paper https://t.ly/AOXru
๐Project https://ai.meta.com/llama
๐GenAI, #Meta unveils Llama-2: a collection of LLMs ranging in scale 7-70B params. Challenging with #chatgpt, but open.
๐Review https://t.ly/bLJgP
๐Paper https://t.ly/AOXru
๐Project https://ai.meta.com/llama
๐คฏ19โค2๐ฅ1๐ฉ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ Controllable Synthetic Data (extending Image-Net) ๐
๐#META's PUG, a new generation of interactive environments for representation learning. Extending Image-Net!
๐Review https://t.ly/nCYs0
๐Paper arxiv.org/pdf/2308.03977.pdf
๐Project pug.metademolab.com
๐Code github.com/facebookresearch/PUG
๐#META's PUG, a new generation of interactive environments for representation learning. Extending Image-Net!
๐Review https://t.ly/nCYs0
๐Paper arxiv.org/pdf/2308.03977.pdf
๐Project pug.metademolab.com
๐Code github.com/facebookresearch/PUG
๐ฅ4โค2๐1๐คฉ1
This media is not supported in your browser
VIEW IN TELEGRAM
โบFACET: Fairness in Computer Visionโบ
๐#META AI opens a large, publicly available dataset for classification, detection & segmentation. Potential performance disparities & challenges across sensitive demographic attributes
๐Review https://t.ly/mKn-t
๐Paper arxiv.org/pdf/2309.00035.pdf
๐Dataset https://facet.metademolab.com/
๐#META AI opens a large, publicly available dataset for classification, detection & segmentation. Potential performance disparities & challenges across sensitive demographic attributes
๐Review https://t.ly/mKn-t
๐Paper arxiv.org/pdf/2309.00035.pdf
๐Dataset https://facet.metademolab.com/
๐ฅ10โค6๐4๐1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ฅ๐ฅ #META's DINOv2 is now commercial! ๐ฅ๐ฅ
๐Universal features for image classification, instance retrieval, video understanding, depth & semantic segmentation. Now suitable for commercial.
๐Review https://t.ly/LNrGy
๐Paper arxiv.org/pdf/2304.07193.pdf
๐Code github.com/facebookresearch/dinov2
๐Demo dinov2.metademolab.com/
๐Universal features for image classification, instance retrieval, video understanding, depth & semantic segmentation. Now suitable for commercial.
๐Review https://t.ly/LNrGy
๐Paper arxiv.org/pdf/2304.07193.pdf
๐Code github.com/facebookresearch/dinov2
๐Demo dinov2.metademolab.com/
๐ฅ15๐3โค1๐คฏ1๐ฑ1
This media is not supported in your browser
VIEW IN TELEGRAM
โ๏ธ Relighted 3D Hands ๐ค
๐#META unveils Re:InterHand: a large dataset of relighted 3D interacting hands
๐Review https://t.ly/I1dQk
๐Paper arxiv.org/pdf/2310.17768.pdf
๐Project mks0601.github.io/ReInterHand
๐Data github.com/mks0601/ReInterHand
๐#META unveils Re:InterHand: a large dataset of relighted 3D interacting hands
๐Review https://t.ly/I1dQk
๐Paper arxiv.org/pdf/2310.17768.pdf
๐Project mks0601.github.io/ReInterHand
๐Data github.com/mks0601/ReInterHand
๐คฏ8โค1๐ฑ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ Emu: image edit / video gen. ๐
๐#Meta the new SOTA in text-to-video generation and instruction-based image editing
๐ Review https://t.ly/PMTBc
๐ Paper (images): https://lnkd.in/eVadH-QS
๐ Project https://lnkd.in/eG8eWUJY
๐ Paper (video): https://lnkd.in/eVadH-QS
๐ Project https://lnkd.in/eu6Zu6gp
๐#Meta the new SOTA in text-to-video generation and instruction-based image editing
๐ Review https://t.ly/PMTBc
๐ Paper (images): https://lnkd.in/eVadH-QS
๐ Project https://lnkd.in/eG8eWUJY
๐ Paper (video): https://lnkd.in/eVadH-QS
๐ Project https://lnkd.in/eu6Zu6gp
๐ฅ8๐คฏ2๐1๐ฑ1
This media is not supported in your browser
VIEW IN TELEGRAM
๐ชRT Humanoid from Head-Mounted Sensors๐ช
๐#META (+CMU) announced SimXR, a method for controlling a simulated avatar from info obtained from AR/VR headsets
๐Review https://t.ly/Si2Mp
๐Paper arxiv.org/pdf/2403.06862.pdf
๐Project www.zhengyiluo.com/SimXR/
๐#META (+CMU) announced SimXR, a method for controlling a simulated avatar from info obtained from AR/VR headsets
๐Review https://t.ly/Si2Mp
๐Paper arxiv.org/pdf/2403.06862.pdf
๐Project www.zhengyiluo.com/SimXR/
โค12โก1๐1
This media is not supported in your browser
VIEW IN TELEGRAM
๐งคHOT3D Hand/Object Tracking๐งค
๐#Meta opens a novel egocentric dataset for 3D hand & object tracking. A new benchmark for vision-based understanding of 3D hand-object interactions. Dataset available ๐
๐Review https://t.ly/cD76F
๐Paper https://lnkd.in/e6_7UNny
๐Data https://lnkd.in/e6P-sQFK
๐#Meta opens a novel egocentric dataset for 3D hand & object tracking. A new benchmark for vision-based understanding of 3D hand-object interactions. Dataset available ๐
๐Review https://t.ly/cD76F
๐Paper https://lnkd.in/e6_7UNny
๐Data https://lnkd.in/e6P-sQFK
๐ฅ9โค3๐3๐2๐คฏ1