Latent RANSAC
We present a method that can evaluate a RANSAC hypothesis in constant time, i.e. independent of the size of the data. A key observation here is that correct hypotheses are tightly clustered together in the latent parameter domain. In a manner similar to the generalized Hough transform we seek to find this cluster, only that we need as few as two votes for a successful detection. Rapidly locating such pairs of similar hypotheses is made possible by adapting the recent ‘Random Grids’ range-search technique. We only perform the usual (costly) hypothesis verification stage upon the discovery of a close pair of hypotheses. We show that this event rarely happens for incorrect hypotheses, enabling a significant speedup of the RANSAC pipeline. The suggested approach is applied and tested on three robust estimation problems: camera localization, 3D rigid alignment and 2D-homography estimation. We perform rigorous testing on both synthetic and real datasets, demonstrating an improvement in efficiency without a compromise in accuracy. Furthermore, we achieve state-of-the-art 3D alignment results on the challenging ‘Redwood’ loop-closure challenge. …

Hierarchically Self Decomposing CNN
Conventional Convolutional neural networks (CNN) are trained on large domain datasets, and are hence typically over-represented and inefficient in limited class applications. An efficient way to convert such large many-class pre-trained networks into small few-class networks is through a hierarchical decomposition of its feature maps. To alleviate this issue, we propose an automated framework for such decomposition in Hierarchically Self Decomposing CNNs (HSD-CNN), in four steps. HSD-CNNs are derived automatically using a class specific filter sensitivity analysis that quantifies the impact of specific features on a class prediction. The decomposed and hierarchical network can be utilized and deployed directly to obtain sub-networks for subset of classes, and it is shown to perform better without the requirement of retraining these sub-networks. Experimental results show that HSD-CNNs generally do not degrade accuracy if the full set of classes are used. However, when operating on known subsets of classes, HSD-CNNs lead to an increased accuracy using a much smaller model size, requiring much less operations. HSD-CNN flow is verified on the CIFAR10, CIFAR100 and CALTECH101 data sets. We report accuracies up to $85.6\%$ ( $94.75\%$ ) on scenarios with 13 ( 4 ) classes of CIFAR100, using a VGG-16 network pretrained on the full data set. In this case, the used HSD-CNN requires $3.97 \times$ fewer parameters and $3.56 \times$ fewer operations than the VGG-16 baseline containing features for all 100 classes. …