Jihong Ju 🤓

4 days until #ICCV2023! Find out all about our #XR demos and come experience them at the @Qualcomm booth next week: "Depth-Guided Neural 3D Scene Reconstruction" and "Improved Self-supervised Depth Estimation on Reflective Surfaces" spaces.qualcomm.com/accelerating-t…

Exited to share our recent paper on depth-guided neural reconstruction this week at #ICCV2023 A shorty summary and link to the paper: spaces.qualcomm.com/accelerating-t… Visit us at the poster or at the Qualcomm booth for more fantastic jobs from Qualcomm XR and AI research!

the TRUTH……

Let's continue working hand in hand on a smooth air bridge, to ensure timely transportations of more protective equipments from China to the Netherlands.@mjrijn #AlleenSamen 🇨🇳🇳🇱

China has banned illegal hunting, trading, transporting and consumption of wildlife. In China, farmers' markets and markets selling live poultry and seafood are not for wildlife trading, and it's actually illegal to do so. Any violations in this regard will be brought to justice.

Farmers are protesting in the Netherlands and it is now the biggest traffic jam we've ever seen, thousands of tractors are driving on the highways right now 😂 #boerenprotest is even worldwide trending on twitter right now🚜🚜🚜

“Conditional Random Field, the factor graph representation” by @jujihong link.medium.com/xEIIYdBvIZ

@UtahMomsLife From one TomTom to another, we're so happy you're home!

“Maximize Likelihood / Maximize Entropy duality” by @jujihong link.medium.com/OmjyKoDaEZ

“Rao Blackwellization in Particle Filter” by @jujihong link.medium.com/bMke2TxpqZ

"Transformer: A Novel Neural Network Architecture for Language Understanding" #coolstuff #feedly research.googleblog.com/2017/08/transf…

PyTorch v0.2 release sooo goood github.com/pytorch/pytorc… more numpy-like (broadcasting/indexing), distributed training, also like retain_grad

"Self-Normalizing Neural Networks" introduced SELUs gitxiv.com/posts/zSmzGXTe…

Memory usage. imatge-upc.github.io/telecombcn-201…

CNTK is adding Keras support. github.com/fchollet/keras…

An Overview of Multi-Task Learning for Deep Learning #fullyadaptivefeaturesharing" target="_blank" rel="nofollow noopener">sebastianruder.com/multi-task/#fu…

The nvidia-378 driver seems not to work with Ubuntu 16.04.2 LTS (kernel based on Linux 4.8). #nvidia #deeplearning

Great animations of Convolution Layers. github.com/vdumoulin/conv…

Mini-batch at pixel level for semantic image segment. introduce sample diversity, fast and computing efficient arxiv.org/pdf/1702.06506…