The Israeli innovation ecosystem is on a roll, earning accolades left and right. After two Israeli startups took home gold in two different categories at the 2020 XTC Global Challenge, the International Conference on Machine Learning (ICML) granted the ‘Outstanding Paper Award’ to Israeli researchers from Stanford University, Israel-based Bar Ilan University, and American Technology giant NVIDIA for their paper, On Learning Sets of Symmetric Elements. The research was led by Hagai Meron from Nvidia's research group in Israel, with Or Litani from Stanford, Ethan Fetaya of Bar Ilan University and Prof. Gal Chechik of Bar Ilan University and the director of the Nvidia Research Group in Israel.
“Our proposed solution is extremely easy to implement, and can be easily integrated with existing network architectures,” the researchers stated.
The work introduces a principled approach to learning sets of general symmetric elements that can be used in a variety of applications, including deblurring image bursts to multi-view 3D shape recognition and reconstruction.
“Our research proves theoretically what deep neural network architectures should be used when learning across sets of complex objects, where by complex we mean that the objects assume a special structure which we refer to as symmetry,” the researchers stated in their paper. “We also show empirically that this architecture achieves superior results in a range of problems over images, graphs, and 3D point clouds.”
The researchers explain that their architectures can be used to reduce noise and identify action highlights given a set of images. One example the researchers tried to solve is identifying the best image in an unordered photo collection of the same scene.
“There are two types of symmetries in this problem: First, the best image should be selected regardless of the collection order. Second, the best image should be selected even if the location of the key elements has shifted slightly.
As it turns out, similar structures are present when working with sets of sounds, signals, images, 3D point-clouds, and even networks. In all of these cases, the elements are ordered arbitrarily, and each element has a special symmetric structure,” the researchers explained. All experiments were conducted using NVIDIA DGX systems with NVIDIA V100 GPUs.