Sound of Pixels: a network learning correspondences between image regions and sound components by watching unlabeled videos. http://sound-of-pixels.csail.mit.edu/
Cool work by Antonio Torralba's group! #CVPR2019

✴️ @AI_Python_EN

Camera localization techniques for AR require persistent storage of digital 3D maps. But deep neural networks can reconstruct detailed images of scenes from such maps. Our solution keeps 3D maps confidential while accurately computing camera pose https://aka.ms/AA5bu2n #CVPR2019
✴️ @AI_Python_EN

paper on Hybrid Task Cascade for Instance Segmentation, ranking 1st in COCO 2018 Challenge Object Detection task.
Project page: http://mmlab.ie.cuhk.edu.hk/projects/HybridTaskCascade/
Code: https://github.com/open-mmlab/mmdetection

✴️ @AI_Python_EN

Dr. Andrew Fitzgibbon is an expert in 3D #computervision and graphics. Discover work on body- and hand-tracking for tech like Kinect and HoloLens and hear how research on dolphins helped build models for the human hand:
https://aka.ms/AA5b1q9 #CVPR2019
✴️ @AI_Python_EN

hierarchical localization paper won the visual localization challenge at #CVPR2019
Paper: https://arxiv.org/abs/1812.03506

✴️ @AI_Python_EN

Facebook & Partnership AI are organizing the 1st Computer Vision for Global Challenges workshop #CVPR2019. they want to help build partnerships between researchers and humanitarian orgs, and discuss how AI can advance the UN sustainable development goals. https://research.fb.com/computer-vision-and-global-challenges-new-research-and-applications/

✴️ @AI_Python_EN

Facebook AI:
Researchers have created 2.5D visual sound by injecting spatial information contained in video frames that accompany a typical monaural audio stream. We've open sourced our data set & videos w/ binaural audio are included. We'll present this at #CVPR2019.
https://ai.facebook.com/blog/visual-sound/

✴️ @AI_Python_EN

When in doubt, people ask for help. What if our personal digital assistants could do the same? Microsoft researchers have created a novel method of training agents to strategically ask for assistance during vision-language tasks:
https://aka.ms/AA5auc5 #CVPR2019

✴️ @AI_Python_EN

Introducing Text2Scene, an interpretable compositional text-to-image synthesis approach https://arxiv.org/abs/1809.01110 // No GANs! but results as good or superior to GANs when it comes to generating scenes.

✴️ @AI_Python_EN

Keyword presence in cvpr2019 paper titles:
- 'Deep' decreasing (taken for granted?).
- 'GAN' saturating.
What's the new 🔥 keyword we should be checking?
#CVPR2019
✴️ @AI_Python_EN

How to make a pizza: Learning a compositional layer-based GAN model. Or “MIT’s AI learns to Become Pizza Guru. All pizza design will soon be automated. ”
https://arxiv.org/abs/1906.02839
#gan #ai #computervision

✴️ @AI_Python_EN

You can find ⁦Women in Computer Vision Workshop papers here from #CVPR2019
http://openaccess.thecvf.com/content_CVPRW_2019/html/WiCV/
✴️ @AI_Python_EN

Creating xBD: A Dataset for Assessing Building Damage from Satellite Imagery" http://bit.ly/xview2-dataset

✴️ @AI_Python_EN

#ICML2019 live from Long Beach, CA, via icmlconf Learn more
→ https://mld.ai/icml2019-live #machinelearning #ML #mldcmu #ICML

✴️ @AI_Python_EN

deep learning for breast cancer screening at the AI for Social Good Workshop at #ICML2019

Paper: https://arxiv.org/abs/1903.08297
Code: https://github.com/nyukat/breast_cancer_classifier

✴️ @AI_Python_EN

Adaptive Neural Trees (ANTs)
Microsoft Research ,We aimed to combine the benefits of decision trees and deep neural networks.
Paper: http://proceedings.mlr.press/v97/tanno19a.html
Code: https://github.com/rtanno21609/AdaptiveNeuralTrees

✴️ @AI_Python_EN

Notes from Thirty-sixth International Conference on Machine Learning here:
https://david-abel.github.io/notes/icml_2019.pdf
#ICML2019

✴️ @AI_Python_EN

Best paper award at #ICML2019 main idea: unsupervised learning of disentangled representations is fundamentally impossible without inductive biases. Verified theoretically & experimentally.
https://arxiv.org/pdf/1811.12359.pdf

✴️ @AI_Python_EN

✴️ @AI_Python_EN

Do you want to improve your generator for free? Just output low energy samples (i.e. filter them with the discriminator)! «Metropolis-Hastings GANs»

✴️ @AI_Python_EN

I'll be sharing 5 Lessons Learned Helping 200,000 non-ML experts* use ML as an #ICML2019 AutoML workshop keynote
https://sites.google.com/view/automl2019icml/schedule?authuser=0

✴️ @AI_Python_EN