Yissum - Research Development Company of the Hebrew University posted this:SPADE: Sub-Image Anomaly Detection with Deep Pyramid CorrespondencesNearest neighbor (kNN) methods utilizing deep pre-trained features exhibit very strong anomaly detection performance when applied to entire images. A limitation of kNN methods is the lack of segmentation map describing where the anomaly lies inside the image. In this work we present a novel anomaly segmentation approach based on alignment between the anomalous image and a constant number of the nearest normal images. Our method, Semantic Pyramid Anomaly Detection (SPADE) uses correspondences based on a multi-resolution feature pyramid. SPADE is shown to achieve state-of-the-art performance on unsupervised anomaly detection and localization while requiring virtually no training time. Project ID : 47-2020-10918
Yissum - Research Development Company of the Hebrew University posted this:Eliminate Blurry Images Caused by Camera Shake and Movementa new architecture which breaks the deblurring network into an analysis network which estimates the blur, and a synthesis network that uses this kernel to deblur the image. Project ID : 47-2020-10908
Yissum - Research Development Company of the Hebrew University posted this:Ultra-Low Dose Radiation for Repeat CT ScanningUltra-Low Dose Radiation for Repeat CT Scanning Project ID : 46-2019-10755
Yissum - Research Development Company of the Hebrew University posted this:Learning from Rough Group Comparisonsin applying machine learning algorithms to the domains with the limited amount of labeled data - cyber, insurance. Developing a machine learning model requires data samples containing both normal and abnormal events, where each sample is labeled accordingly; however,it is often the case that unlabeled data is abundant, but labeled instances are difficult, expensive, or time-consuming to obtain. We proposed a simple and intuitive bi-convex formulation and an efficient algorithm for the problem. Despite the simplicity of our methods, we achieve high accuracy with a very modest amount of input, and considerably loose (or even misspecified) constraints Project ID : 46-2020-13911
Arkadiusz Rudawski, PhD posted this:
Technology Broker at Cracow University of Technology
Universitat Politècnica de Catalunya - UPC posted this:Method and system for correcting errors in satellite positioning systems and computer program products thereofAn innovative approach to improve the performance of tablet and smartphone embedded GNSS positioning services. Partners to develop the system and/or to establish commercial agreements along with technical cooperation are sought.
Yissum - Research Development Company of the Hebrew University posted this:Takeover Detection for Cyber Defense at the Industrial Control Systemsdetection of a cyber takeover of the monitored Industrial Control Systems and critical infrastructures, even in the presence of successful sensor hijacking Project ID : 46-2019-10804
Yissum - Research Development Company of the Hebrew University posted this:Repurposing of a Natural Compound for Treatment for Neuropathic PainRepurposing of a Natural Compound for Treatment for Neuropathic Pain Project ID : 6-2019-9749
Yissum - Research Development Company of the Hebrew University posted this:Early Detection and Personalized Treatment of Breast Cancer, Using Comparative GenomicsUsing cutting-edge comparative genomics, data integration and machine learning, the researchers will develop a comprehensive gene panel that can identify increased risk of breast cancer and offer personalized treatment options based on PARP inhibitors Project ID : 6-2018-6690
Center for Technology Transfer and Commercialization of Novosibirsk State University posted this:New class of computing devices with memory based on graphene nanocompositesModels of mem-capacitors - computing devices with memory based on graphene nanocomposites are developed in the laboratory of the university (Russia). This class of devices has an architecture with significantly greater speed and lower power consumption compared to traditional ones. It allows avoiding the memory-processor bandwidth problem of the device. The laboratory is looking for partners for technical cooperation, licensing or commercial agreements with technical assistance.
Center for Technology Transfer and Commercialization of Novosibirsk State University posted this:Post-quantum cryptosystem with public keyA post-quantum public-key cryptographic system has been developed in the cryptography laboratory of the university (Novosibirsk). It is used in the Internet of things, blockchain technology, cryptocurrency, key exchange, for example, in instant messengers, or as an independent mean of protection. The cryptosystem is resistant to hacking, including a quantum computer through the use of NP-hard in a quantum sense problem. The university is looking for partners for furtherance on the Russian and international markets.
Yissum - Research Development Company of the Hebrew University posted this:Audio Speech Enhancement by Computer Vision AnalyticsCluster4 Cluster5 This was filed due to a meeting that Shmuel had with Orcam Project ID : 10-2018-4602
Yissum - Research Development Company of the Hebrew University posted this:Method for Smart, Unbiased Matching Between Job Seekers and RecruitersCluster3 Based on Gati's screening and recommendation program, intended for later stage career decision (not just what to study). The emphasis is more on the recommendation stage and helping to choose from a few alternatives based on past experiences and various considerations (more so than the initial stages of screening and deep investigation). Project ID : 25-2018-4557
Yissum - Research Development Company of the Hebrew University posted this:Nanoparticle Sensor for RDX and Other Nitroamine-Based ExplosivesNanoparticles matrix for the detection of non-planar and non-aromatic nitroamine analytes. Keywords: explosive detection, RDX, nitroamines, nanosensor, Devices, Molecular Electronics, Nano Materials and Nano Structures, Soil, Pollutant Detection, Homeland Security Homeland Security Project ID : 9-2010-2364