Cristina Ferrero Castaño
A Year Full of Amazing AI Papers - A Review [Part Two]
A curated list of the latest breakthroughs in AI by release date with a clear video explanation, link to a more in-depth article, and code
Medium
Louis (What´s AI) Bouchard
Dec 21, 2020
In this second part we will continue discovering the most interesting research papers of 2020...
Even with everything that happened in the world this year, we still had the chance to see a lot of amazing research come out. Especially in the field of artificial intelligence. More, many important aspects were highlighted this year, like the ethical aspects, important biases, and much more. Artificial intelligence and our understanding of the human brain and its link to AI is constantly evolving, showing promising applications in the soon future.
Here are the most interesting research papers of the year, in case you missed any of them. In short, it is basically a curated list of the latest breakthroughs in AI and Data Science by release date with a clear video explanation, link to a more in-depth article, and code (if applicable).
...
[15] I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image
Their goal was to propose a new technique for 3D Human Pose and Mesh Estimation from a single RGB image. They called it I2L-MeshNet. Where I2L stands for Image-to-Lixel. Just like a voxel, volume + pixel, is a quantized cell in three-dimensional space, they defined lixel, a line, and pixel, as a quantized cell in one-dimensional space. Their method outperforms previous methods and the code is publicly available!
[16] Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environments
Language-guided navigation is a widely studied field and a very complex one. Indeed, it may seem simple for a human to just walk through a house to get to your coffee that you left on your nightstand to the left of your bed. But it is a whole other story for an agent, which is an autonomous AI-driven system using deep learning to perform tasks.
[17] RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
ECCV 2020 Best Paper Award Goes to Princeton Team. They developed a new end-to-end trainable model for optical flow. Their method beats state-of-the-art architectures’ accuracy across multiple datasets and is way more efficient. They even made the code available for everyone on their Github!
[18] Crowdsampling the Plenoptic Function
Using tourists’ public photos from the internet, they were able to reconstruct multiple viewpoints of a scene conserving the realistic shadows and lighting! This is a huge advancement of the state-of-the-art techniques for photorealistic scene rendering and their results are simply amazing.
[19] Old Photo Restoration via Deep Latent Space
Translation
Imagine having the old, folded, and even torn pictures of your grandmother when she was 18 years old in high definition with zero artifacts. This is called old photo restoration and this paper just opened a whole new avenue to address this problem using a deep learning approach.
[20] Neural circuit policies enabling auditable autonomy
Researchers from IST Austria and MIT have successfully trained a self-driving car using a new artificial intelligence system based on the brains of tiny animals, such as threadworms. They achieved that with only a few neurons able to control the self-driving car, compared to the millions of neurons needed by the popular deep neural networks such as Inceptions, Resnets, or VGG. Their network was able to completely control a car using only 75 000 parameters, composed of 19 control neurons, rather than millions!
[21] Lifespan Age Transformation Synthesis
A team of researchers from Adobe Research developed a new technique for age transformation synthesis based on only one picture from the person. It can generate the lifespan pictures from any picture you sent it.
[22] DeOldify
DeOldify is a technique to colorize and restore old black and white images or even film footage. It was developed and is still getting updated by only one person Jason Antic. It is now the state of the art way to colorize black and white images, and everything is open-sourced, but we will get back to this in a bit.
[23] COOT: Cooperative Hierarchical Transformer for Video-Text Representation Learning
As the name states, it uses transformers to generate accurate text descriptions for each sequence of a video, using both the video and a general description of it as inputs.
[24] Stylized Neural Painting
This Image-to-Painting Translation method simulates a real painter on multiple styles using a novel approach that does not involve any GAN architecture, unlike all the current state-of-the-art approaches!
[25] Is a Green Screen Really Necessary for Real-Time Portrait Matting?
Human matting is an extremely interesting task where the goal is to find any human in a picture and remove the background from it. It is really hard to achieve due to the complexity of the task, having to find the person or people with the perfect contour. In this post, I review the best techniques used over the years and a novel approach published on November 29th, 2020. Many techniques are using basic computer vision algorithms to achieve this task, such as the GrabCut algorithm, which is extremely fast, but not very precise.
[26] ADA: Training Generative Adversarial Networks with Limited Data
With this new training method developed by NVIDIA, you can train a powerful generative model with one-tenth of the images! Making possible many applications that do not have access to so many images!
[27] Improving Data‐Driven Global Weather Prediction Using Deep Convolutional Neural Networks on a Cubed Sphere
The current traditional approach for weather forecasting uses what we call “Numerical weather prediction” models. It uses mathematical models of the atmosphere and oceans to predict the weather based on the current conditions. It was first introduced in the 1920s and produced realistic results in the 1950s using computer simulations. These mathematical models work for predicting both short and long-term forecasts. But it’s heavy in computation and cannot base its predictions on as much data as a deep neural network. This is partly why it is so promising. These current numerical weather prediction models already use machine learning to improve the forecasts as a post-processing tool. Weather forecasting is receiving more and more attention from machine learning researchers, already yielding promising results.
[28] NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesis
This new method is able to generate a complete 3-dimensional scene and has the ability to decide the lighting of the scene. All this with very limited computation costs and amazing results compared to previous approaches.
To access the full article where you can find more detailed articles and the different codes...