Man-made scenes can be densely packed, containing numerous objects, often identical, positioned in close proximity. We show that precise object detection in such scenes remains a challenging frontier even for state-of-the-art object detectors. We propose a novel, deep-learning based method for precise object detection, designed for such challenging settings. Our contributions include: (1) A layer for estimating the Jaccard index as a detection quality score;(2) a novel EM merging unit, which uses our quality scores to resolve detection overlap ambiguities; finally, (3) an extensive, annotated data set, SKU-110K, representing packed retail environments, released for training and testing under such extreme settings. Detection tests on SKU-110K and counting tests on the CARPK and PUCPR+ show our method to outperform existing state-of-the-art with substantial margins. The code and data will be made available on

No abstract

No abstract

No abstract

We describe a system that automatically acquires a language model for a particular task from semantic-level information. This is in contrast to systems with predefined vocabulary and syntax. The purpose of the system is to map spoken or typed input into a machine action. To accomplish this task we use a medium-grain neural network. We introduce a novel adaptive training procedure for estimating the connection weights, which has the advantages of rapid, single-pass and orderinvariant learning.The resulting weights have information-theoretic significence, and do not require gradient search techniques for their estimation.We experimentally evaluate the system on three text-based tasks. The first is a three-class inward-call manager with an acquired vocabulary of over 1600 words. The second is a fifteen-action subset of the DARPA Resource Manager, with an acquired vocabulary of over 700 words. The third is to discriminate between idiomatic phrases meaning 'yes' or 'no'.