ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
📌 Project: https://opennlplab.github.io/AVSBench/
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
Forwarded from Machinelearning
Over 3000 models, and over 100 datasets on the Hugging Face Hub.
Более 3000 моделей компьютерного зрения и более 100 датасетов на Hugging Face Hub.
Supported vision tasks and Pipelines
Training your own vision models
Integration with timm
Diffusers
Support for third-party libraries
Datasets
HugsVision
Model documentation
Hugging Face notebooks
Hugging Face example scripts
Task pages
Timm
Generate 3D voxels from a predicted depth map of an input image
Open vocabulary semantic segmentation
Narrate videos by generating captions
Classify videos from YouTube
Zero-shot video classification
Visual question-answering
Use zero-shot image classification to find best captions for an image to generate similar images
🤗 AutoTrain
AutoTrain
Image classification
Automatic model evaluation
🦾 Zero-shot models
CLIP
OWL-ViT
CLIPSeg
GroupViT
X-CLIP
🚀 Deployment
Deploying TensorFlow Vision Models in Hugging Face with TF Serving
Deploying ViT on Kubernetes with TF Serving
Deploying ViT on Vertex AI
Deploying ViT with TFX and Vertex AI
@ai_machinelearning_big_data
Please open Telegram to view this post
VIEW IN TELEGRAM
NP-Match: When Neural Processes meet Semi-Supervised Learning
🖥 Github: https://github.com/jianf-wang/np-match
⏩ Paper: https://arxiv.org/abs/2301.13569v1
➡️ Dataset: https://paperswithcode.com/dataset/stl-10
ArtificialIntelligencedl
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
What Makes Good Examples for Visual In-Context Learning?
🖥 Github: https://github.com/zhangyuanhan-ai/visual_prompt_retrieval
⏩ Paper: https://arxiv.org/abs/2301.13670v2
➡️ Dataset: https://paperswithcode.com/dataset/coco
ArtificialIntelligencedl
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
Forwarded from Machinelearning
This media is not supported in your browser
VIEW IN TELEGRAM
🔥 Dreamix: Video Diffusion Models are General Video Editors
New Google's text-based motion model.
Given a small collection of images showing the same subject, Dreamix can generate new videos with the subject in motion.
Всего из нескольких картинок или ролику новая модель от Google - Dreamix генерирует видео по текстовому описанию!
На видео Dreamix превращает обезьяну в танцующего медведя по промпту «Медведь танцует и прыгает под веселую музыку, двигая всем телом».
⭐️ Project: https://dreamix-video-editing.github.io/
✅️ Paper: https://arxiv.org/pdf/2302.01329.pdf
⭐️ Video: https://www.youtube.com/watch?v=xcvnHhfDSGM
ai_machinelearning_big_data
New Google's text-based motion model.
Given a small collection of images showing the same subject, Dreamix can generate new videos with the subject in motion.
Всего из нескольких картинок или ролику новая модель от Google - Dreamix генерирует видео по текстовому описанию!
На видео Dreamix превращает обезьяну в танцующего медведя по промпту «Медведь танцует и прыгает под веселую музыку, двигая всем телом».
ai_machinelearning_big_data
Please open Telegram to view this post
VIEW IN TELEGRAM
STEPS: Joint Self-supervised Nighttime Image Enhancement and Depth Estimation (ICRA 2023)
🖥 Github: https://github.com/ucaszyp/steps
⏩ Paper: https://arxiv.org/abs/2302.01334v1
➡️ Dataset: https://paperswithcode.com/dataset/nuscenes
ArtificialIntelligencedl
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
This media is not supported in your browser
VIEW IN TELEGRAM
DirectMHP: Direct 2D Multi-Person Head Pose Estimation
🖥 Github: https://github.com/hnuzhy/directmhp
⏩ Paper: https://arxiv.org/abs/2302.01110v1
➡️ Dataset: https://paperswithcode.com/dataset/agora
ArtificialIntelligencedl
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
OpenSpike: An OpenRAM SNN Accelerator
🖥 Github: https://github.com/sfmth/openspike
⏩ Paper: https://arxiv.org/abs/2302.01015v1
ArtificialIntelligencedl
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
Multimodal Chain-of-Thought Reasoning in Language Models
🖥 Github: https://github.com/amazon-science/mm-cot
⏩ Paper: https://arxiv.org/abs/2302.00923v1
➡️ Dataset: https://paperswithcode.com/dataset/scienceqa
ArtificialIntelligencedl
ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class-Incremental Learning
🖥 Github: https://github.com/NeuralCollapseApplications/FSCIL
⏩ Paper: https://openreview.net/pdf?id=y5W8tpojhtJ
➡️ Dataset: https://paperswithcode.com/dataset/mini-imagenet
ArtificialIntelligence
ArtificialIntelligence
Please open Telegram to view this post
VIEW IN TELEGRAM
Tab2KG: Semantic Table Interpretation with Lightweight Semantic Profiles
🖥 Github:https://github.com/sgottsch/tab2kg
⏩ Paper: https://arxiv.org/pdf/2302.01150v1.pdf
➡️ Dataset: https://paperswithcode.com/dataset/dbpedia
ArtificialIntelligence
🖥 Github:https://github.com/sgottsch/tab2kg
⏩ Paper: https://arxiv.org/pdf/2302.01150v1.pdf
➡️ Dataset: https://paperswithcode.com/dataset/dbpedia
ArtificialIntelligence
Top-Down Beats Bottom-Up in 3D Instance Segmentation
🖥 Github: https://github.com/samsunglabs/td3d
⏩ Paper: https://arxiv.org/abs/2302.02871v1
➡️ Dataset: https://paperswithcode.com/dataset/scannet
ArtificialIntelligence
ArtificialIntelligence
Please open Telegram to view this post
VIEW IN TELEGRAM
Expert Language Models (ELM)
🖥 Github: https://github.com/joeljang/elm
⏩ Paper: https://arxiv.org/abs/2302.03202v1
➡️ Dataset: https://paperswithcode.com/dataset/big-bench
ArtificialIntelligence
ArtificialIntelligence
Please open Telegram to view this post
VIEW IN TELEGRAM
Сбер решил сделать подарок всем любителям научного знания и запустил сайт ко Дню российской науки. На сайте можно найти информацию об исследованиях и разработках (R&D) Сбербанка за последние годы.
Также там можно почитать об открытиях и проектах 10 лабораторий Сбера по топовым направлениям науки, в частности:
➡️блокчейн
➡️нейронауки
➡️AR/VR
➡️геймификация
➡️интернет вещей
➡️кибербезопасность
➡️искусственный интеллект
Кроме того можно узнать о партнёрских проектах лабораторий Сбера с ведущими вузами страны и центрами искусственного интеллекта на базе ВШЭ, Сколтеха и МФТИ, присоединиться к мероприятиям, которые проводят исследователи Сбера.
Кому мало, могут посетить специальный проект для всех, кто интересуется наукой.
ArtificialIntelligencedl
Также там можно почитать об открытиях и проектах 10 лабораторий Сбера по топовым направлениям науки, в частности:
➡️блокчейн
➡️нейронауки
➡️AR/VR
➡️геймификация
➡️интернет вещей
➡️кибербезопасность
➡️искусственный интеллект
Кроме того можно узнать о партнёрских проектах лабораторий Сбера с ведущими вузами страны и центрами искусственного интеллекта на базе ВШЭ, Сколтеха и МФТИ, присоединиться к мероприятиям, которые проводят исследователи Сбера.
Кому мало, могут посетить специальный проект для всех, кто интересуется наукой.
ArtificialIntelligencedl
NASiam: Efficient Representation Learning using Neural Architecture Search for Siamese Networks
🖥 Github:https://github.com/aheuillet/nasiam
⏩ Paper: https://arxiv.org/pdf/2302.00059v1.pdf
➡️ Dataset:https://paperswithcode.com/dataset/cifar-100
ArtificialIntelligenceу
🖥 Github:https://github.com/aheuillet/nasiam
⏩ Paper: https://arxiv.org/pdf/2302.00059v1.pdf
➡️ Dataset:https://paperswithcode.com/dataset/cifar-100
ArtificialIntelligenceу
FakeSV: A Multimodal Benchmark with Rich Social Context for Fake News Detection on Short Video Platforms
🖥 Github: https://github.com/ictmcg/fakesv
⏩ Paper: https://arxiv.org/abs/2302.03242v1
➡️ Data Processing: https://github.com/YaoFANGUK/video-subtitle-extractor
ArtificialIntelligenceу
ArtificialIntelligenceу
Please open Telegram to view this post
VIEW IN TELEGRAM
This media is not supported in your browser
VIEW IN TELEGRAM
PySlowFast
🖥 Github: https://github.com/facebookresearch/SlowFast
⏩ Paper: https://arxiv.org/abs/2302.04869v1
➡️ Dataset: https://paperswithcode.com/dataset/kinetics-400-1
ArtificialIntelligenceу
ArtificialIntelligenceу
Please open Telegram to view this post
VIEW IN TELEGRAM
ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skills
🖥 Github: https://github.com/haosulab/maniskill2
🖥 Colab: https://colab.research.google.com/github/haosulab/ManiSkill2/blob/main/examples/tutorials/1_quickstart.ipynb
📎 Docs: https://haosulab.github.io/ManiSkill2
⏩ Paper: https://arxiv.org/abs/2302.04659v1
➡️ Dataset: https://paperswithcode.com/dataset/plasticinelab
ArtificialIntelligenceу
📎 Docs: https://haosulab.github.io/ManiSkill2
ArtificialIntelligenceу
Please open Telegram to view this post
VIEW IN TELEGRAM
Polynomial Neural Fields for Subband Decomposition and Manipulation
🖥 Github: https://github.com/stevenygd/pnf
⏩ Paper: https://arxiv.org/abs/2302.04862v1
➡️ Dataset: https://paperswithcode.com/dataset/div2k
ArtificialIntelligenceу
ArtificialIntelligenceу
Please open Telegram to view this post
VIEW IN TELEGRAM