Showing papers by "Sony Broadcast & Professional Research Laboratories published in 2015"

Midori: A Block Cipher for Low Energy

Abstract: In the past few years, lightweight cryptography has become a popular research discipline with a number of ciphers and hash functions proposed. The designers' focus has been predominantly to minimize the hardware area, while other goals such as low latency have been addressed rather recently only. However, the optimization goal of low energy for block cipher design has not been explicitly addressed so far. At the same time, it is a crucial measure of goodness for an algorithm. Indeed, a cipher optimized with respect to energy has wide applications, especially in constrained environments running on a tight power/energy budget such as medical implants. This paper presents the block cipher Midorii¾?The name of the cipher is the Japanese translation for the word Green. that is optimized with respect to the energy consumed by the circuit per bt in encryption or decryption operation. We deliberate on the design choices that lead to low energy consumption in an electrical circuit, and try to optimize each component of the circuit as well as its entire architecture for energy. An added motivation is to make both encryption and decryption functionalities available by small tweak in the circuit that would not incur significant area or energy overheads. We propose two energy-efficient block ciphers Midori128i¾?and Midori64i¾?with block sizes equal to 128 and 64 bits respectively. These ciphers have the added property that a circuit that provides both the functionalities of encryption and decryption can be designed with very little overhead in terms of area and energy. We compare our results with other ciphers with similar characteristics: it was found that the energy consumptions ofi¾?Midori64i¾? and Midori128i¾? are by far better when compared ciphers like PRINCE and NOEKEON.

Learning scene-specific pedestrian detectors without real data

Abstract: We consider the problem of designing a scene-specific pedestrian detector in a scenario where we have zero instances of real pedestrian data (i.e., no labeled real data or unsupervised real data). This scenario may arise when a new surveillance system is installed in a novel location and a scene-specific pedestrian detector must be trained prior to any observations of pedestrians. The key idea of our approach is to infer the potential appearance of pedestrians using geometric scene data and a customizable database of virtual simulations of pedestrian motion. We propose an efficient discriminative learning method that generates a spatially-varying pedestrian appearance model that takes into the account the perspective geometry of the scene. As a result, our method is able to learn a unique pedestrian classifier customized for every possible location in the scene. Our experimental results show that our proposed approach outperforms classical pedestrian detection models and hybrid synthetic-real models. Our results also yield a surprising result, that our method using purely synthetic data is able to outperform models trained on real scene-specific data when data is limited.

Solid-state imaging device, method of manufacturing the same, and electronic apparatus

Abstract: A solid state imaging device that includes a substrate having oppositely facing first and second surfaces and a photoelectric conversion unit layer having a light incident side facing away from the substrate. The substrate includes a first photoelectric conversion unit and a second photoelectric conversion and the photoelectric conversion layer includes a third photoelectric conversion unit.

Direct Intrinsics: Learning Albedo-Shading Decomposition by Convolutional Regression

Abstract: We introduce a new approach to intrinsic image decomposition, the task of decomposing a single image into albedo and shading components. Our strategy, which we term direct intrinsics, is to learn a convolutional neural network (CNN) that directly predicts output albedo and shading channels from an input RGB image patch. Direct intrinsics is a departure from classical techniques for intrinsic image decomposition, which typically rely on physically-motivated priors and graph-based inference algorithms. The large-scale synthetic ground-truth of the MPI Sintel dataset plays the key role in training direct intrinsics. We demonstrate results on both the synthetic images of Sintel and the real images of the classic MIT intrinsic image dataset. On Sintel, direct intrinsics, using only RGB input, outperforms all prior work, including methods that rely on RGB+Depth input. Direct intrinsics also generalizes across modalities, our Sintel-trained CNN produces quite reasonable decompositions on the real images of the MIT dataset. Our results indicate that the marriage of CNNs with synthetic training data may be a powerful new technique for tackling classic problems in computer vision.

Transmission device, transmission method, reception device, and reception method

Abstract: The purpose of the present invention is to enable correct stream reception to take place at the receiving end, in a manner accurately adapted to dynamic changes in delivered content. In a first transmission mode, a predetermined number of extended video streams constituting scalable-coded image data, and respectively including a basic video stream that includes lowest-layer image data, and image data of layers other than the lowest layer. In a second transmission mode, only the basic video stream that includes basic image data is transmitted. Identifying information for identifying the first mode and the second mode is inserted into the video stream. At the receiving end, the transmission mode can be easily ascertained, making it possible for correct stream reception to take place in a manner accurately adapted to changes in the stream composition, that is, dynamic changes in delivered content.

Physical information acquisition method, physical information acquisition device, and semiconductor device

Abstract: A physical information acquisition method in which a corresponding wavelength region of visible light with at least one visible light detection unit coupled to an image signal processing unit is detected, each said visible light detection unit comprising a color filter adapted to transmit the corresponding wavelength region of visible light; a wavelength region of infrared light with at least one infrared light detection unit coupled to the image signal processing unit is detected; and, with the signal processing unit, a first signal received from the at least one visible light detection unit by subtracting a product from said first signal is corrected, said product resulting from multiplication of a second signal received from the at least one infrared light detection unit and a predetermined coefficient factor.

Deep neural network based instrument extraction from music

Abstract: This paper deals with the extraction of an instrument from music by using a deep neural network. As prior information, we only assume to know the instrument types that are present in the mixture and, using this information, we generate the training data from a database with solo instrument performances. The neural network is built up from rectified linear units where each hidden layer has the same number of nodes as the output layer. This allows a least squares initialization of the layer weights and speeds up the training of the network considerably compared to a traditional random initialization. We give results for two mixtures, each consisting of three instruments, and evaluate the extraction performance using BSS Eval for a varying number of hidden layers.

Direct Intrinsics: Learning Albedo-Shading Decomposition by Convolutional Regression

Abstract: We introduce a new approach to intrinsic image decomposition, the task of decomposing a single image into albedo and shading components. Our strategy, which we term direct intrinsics, is to learn a convolutional neural network (CNN) that directly predicts output albedo and shading channels from an input RGB image patch. Direct intrinsics is a departure from classical techniques for intrinsic image decomposition, which typically rely on physically-motivated priors and graph-based inference algorithms. The large-scale synthetic ground-truth of the MPI Sintel dataset plays a key role in training direct intrinsics. We demonstrate results on both the synthetic images of Sintel and the real images of the classic MIT intrinsic image dataset. On Sintel, direct intrinsics, using only RGB input, outperforms all prior work, including methods that rely on RGB+Depth input. Direct intrinsics also generalizes across modalities; it produces quite reasonable decompositions on the real images of the MIT dataset. Our results indicate that the marriage of CNNs with synthetic training data may be a powerful new technique for tackling classic problems in computer vision.

Solid-state imaging device and camera system

Abstract: A solid-state imaging device and a camera system are disclosed. The solid-state imaging device includes a pixel unit and a pixel signal readout circuit. The pixel signal readout circuit includes a plurality of comparators disposed to correspond to a pixel column array, and a plurality of counters. Each counter includes a first amplifier, a second amplifier, and a mirror circuit to from a current mirror in parallel with the second amplifier. The first amplifier includes differential transistors, initializing switches connected between gates and collectors of the differential transistors, and first and second capacitors connected to each of the gates of the differential transistors. The second amplifier includes an initializing switch and a third capacitor. The mirror circuit includes a gate input transistor whose gate is inputted with a voltage sampled by the first amplifier or a voltage sampled by the second amplifier.

Room-temperature continuous-wave operation of GaN-based vertical-cavity surface-emitting lasers fabricated using epitaxial lateral overgrowth

Abstract: We have successfully demonstrated the room-temperature continuous-wave operation of GaN-based vertical-cavity surface-emitting lasers (VCSELs) with all-dielectric reflectors, which were fabricated using epitaxial lateral overgrowth. The VCSELs exhibited a threshold current of 8 mA and a threshold voltage of 4.5 V at a lasing wavelength of 446 nm. The maximum output power was 0.9 mW for an 8-µm-diameter current aperture, which was made possible because of the high thermal conductivity of the GaN substrate.

Communications device, infrastructure equipment, mobile communications network and methods

Abstract: An infrastructure equipment forming part of a mobile communications network transmits data to or receives data from communications devices. The infrastructure equipment comprises a transmitter configured to transmit the signals to communications devices via a wireless access interface, a receiver configured to receive the signals from the communications devices via the wireless access interface, and a controller for controlling the transmitter and the receiver to form the wireless access interface for transmitting to the communications devices and receiving the data from the communications devices. The wireless access interface providing a first primary carrier within a first frequency range, which forms a primary cell providing a contiguous set of communications resources across the first frequency range. The controller is configured in combination with the receiver and transmitter to receive from one or more of the communications devices a channel state report for each of one or more predefined candidate channels within a second frequency range which is different to and mutually exclusive from the first frequency range, each of the candidate channels in the second frequency range representing a minimum unit of communications resource which can be used to receive data on the downlink. The controller is configured to select, from the one or more candidate carriers, one or more component carriers for use in transmitting signals to the communications devices within the second frequency range to form a down-link providing a secondary cell, based on the received channel state reports.

White-Box Cryptography Revisited: Space-Hard Ciphers

Abstract: The need for software security in untrusted environments is ever increasing. White-box cryptography aims to ensure the security of cryptographic algorithms when the attacker has full access to their implementations. However, there is no secure white-box implementation of standard block ciphers such as DES and AES known to date: All published techniques have been practically broken. In this paper, we revisit white-box cryptography and propose a family of white-box secure block ciphers SPACE with several novel features. The design of SPACE is such that the key-extraction security in the white box reduces to the well-studied problem of key recovery for block ciphers (AES in our example) in the standard black-box setting. Moreover, to mitigate code lifting, we introduce the notion of space hardness. It measures the difficulty of compressing the white-box implementation of a cipher, and quantifies security against code lifting by the amount of code that needs to be extracted from the implementation by a white-box attacker to maintain its functionality. SPACE includes several variants with different white-box code sizes. Therefore, it is applicable to a wide range of environments and use cases. One of the variants called N-SPACE can be implemented with different code sizes while keeping the cipher itself unchanged. SPACE offers a high level of space hardness: It is difficult to find a compact but still functional representation of SPACE given its white-box implementation. This property has several useful consequences for applications. First, it gets more challenging for a DRM attacker (e.g. in a pay TV setting) to scale a code-lifting attack and to distribute the break. Moreover, this paves the way for mass-surveillance resistant cryptography: If a large proportion of users dedicates a significant part of their computers' storage (e.g. HDD) to white-box SPACE implementations, it will be much more complex or even infeasible for governmental agencies to deal with the keys of all users simultaneously due to the limited storage available, forcing them to focus on targeted attacks instead. This consequence is especially important given Snowden's revelations on the extent of the mass surveillance practice by NSA and GCHQ. Finally, the usage of SPACE ciphers can mitigate the damage of having malware in security-critical systems such as networks processing top-secret data: As those are typically insulated from the Internet, the capacity of the communication channel from inside to outside the system is often limited, making it infeasible for Trojans to transmit the necessary key material.

Automatic driving control device and automatic driving control method, and program

Abstract: A method for switching modes for operating a vehicle, a non-transitory computer-readable medium for performing the method, and information processing apparatuses. The method includes determining, by circuitry of an information processing apparatus, whether a mode for operating the vehicle is to be switched from one of autonomous and manual driving modes to the other of the autonomous and manual driving modes. A state of a driver of the vehicle is obtained when the mode for operating the vehicle is determined to be switched. The method further includes switching, by the circuitry, the mode for operating the vehicle from the one of the autonomous and manual driving modes to the other of the autonomous manual driving modes based on the obtained state of the driver.

Light emitting device and display device

Abstract: A light-emitting device of the present disclosure includes: a first light-guiding plate having a first light incidence surface and a first light emission enhancement surface provided with a first pattern; and a second light-guiding plate having a second light incidence surface and a second light emission enhancement surface provided with a second pattern. The first pattern is a pattern constituted by a plurality of first punctiform sections of which an arrangement pitch changes to cause density of the first punctiform sections to increase as a distance from the first light incidence surface increases, and the second pattern is a pattern constituted by a plurality of second punctiform sections of which an arrangement pitch changes to cause density of the second punctiform sections to increase as a distance from the second light incidence surface increases, and when viewed from the light emission direction, regions provided with the first pattern and the second pattern partially overlap, and directions in which densities of the first pattern and the second pattern increase are opposite to each other.

Communication system, a communication method, and a communication apparatus

Abstract: A communication system for carrying out data communication among a plurality of communication stations is disclosed in which a first communication station for transmitting to other communication stations a Request To Send (RTS) signal for requesting a transmission upon the start of the data transmission; and a plurality of second communication stations transmitting to other communication stations a Clear To Send (CTS) signal for notifying the completion of preparing the reception, wherein the first communication station transmits the RTS signal describing at least each of addresses the second communication stations that are desired to receive the data, and receives a plurality of CTS signals transmitted from each of the second communication stations in order to increase communication capacity

Non-volatile memory system with reset verification mechanism and method of operation thereof

Abstract: A method of operation of a non-volatile memory system includes: providing a resistive storage element having a high resistance state and a low resistance state; coupling an analog multiplexer to the resistive storage element for applying a bias voltage; and switching between a verification bias and a read bias through the analog multiplexer for increasing a read margin between the high resistance state and the low resistance state.

Solid-state imaging device, method for processing signal of solid-state imaging device, and imaging apparatus

Abstract: A solid-state imaging device includes a color filter unit disposed on a pixel array unit including pixels two-dimensionally arranged in a matrix and a conversion processing unit disposed on a substrate having the pixel array unit thereon. The color filter unit has a color arrangement in which a color serving as a primary component of a luminance signal is arranged in a checkerboard pattern and a plurality of colors serving as color information components are arranged in the other area of the checkerboard pattern. The conversion processing unit converts signals that are output from the pixels of the pixel array unit and that correspond to the color arrangement of the color filter unit into signals that correspond to a Bayer arrangement and outputs the converted signals.

Image processing apparatus, image processing method, and program

Abstract: Provided is an image processing apparatus including a generation section which generates, based on a plurality of images captured at different exposure times, index information representing an index of a brightness of the plurality of images, and a detection section which detects a moving photographic subject included in the plurality of images based on the index information.

Reception apparatus, reception method, transmission apparatus, and transmission method

Abstract: Enable appropriate caching of application programs executed in coordination with AV content. The cache memory temporarily stores a coordinated application that is executed in coordination with received AV content. When a plurality of coordinated applications are stored in the cache memory, a control block controls the cache memory according to the priority of the coordinated applications which corresponds to a predetermined frequency specified for each of them in advance. The present technology is applicable to, for example, television receivers that receive digital television broadcast signals.

Receiving device, receiving method, program, and broadcasting system

Abstract: Disclosed herein is a receiving device for receiving AV contents, the receiving device including: an obtaining section configured to obtain a description document described in a program language for displaying contents from a server present on a network connected with the receiving device; a determining section configured to determine whether the obtained the description document is obtained from a first server managed on a broadcaster side broadcasting the AV contents; and a controlling section configured to control execution of the obtained the description document according to a result of determination by the determining section.

Sulfone-based electrolytes for aluminium rechargeable batteries

Abstract: Electrolyte is a key material for success in the research and development of next-generation rechargeable batteries. Aluminium rechargeable batteries that use aluminium (Al) metals as anode materials are attractive candidates for next-generation batteries, though they have not been developed yet due to the lack of practically useful electrolytes. Here we present, for the first time, non-corrosive reversible Al electrolytes working at room temperature. The electrolytes are composed of aluminium chlorides, dialkylsulfones, and dilutants, which are realized by the identification of electrochemically active Al species, the study of sulfone dependences, the effects of aluminium chloride concentrations, dilutions and their optimizations. The characteristic feature of these materials is the lower chloride concentrations in the solutions than those in the conventional Al electrolytes, which allows us to use the Al metal anodes without corrosions. We anticipate that the sulfone-based electrolytes will open the doors for the research and development of Al rechargeable batteries.

An Encryption-then-Compression system for JPEG 2000 standard

Abstract: A new Encryption-then-Compression (ETC) system for the JPEG 2000 standard is proposed in this paper. An ETC system is known as a system that makes image communication secure and efficient by using perceptual encryption and image compression. The proposed system uses the sign-scrambling and block-shuffling of discrete wavelet transform (DWT) coefficients as perceptual encryption. Unlike conventional ETC systems, the proposed system is compatible with the JPEG 2000 standard because the perceptually encrypted coefficients can be efficiently compressed by the JPEG 2000. The experimental results demonstrated that the proposed system achieved both acceptable compression performance and enough key-space for secure image communication while remaining compatible with the JPEG 2000 standard.

Vehicle control device, vehicle control method, and program

Abstract: An environment information acquisition unit 20 acquires an image in which environment around a vehicle has been imaged, information indicating a contact with the vehicle or an impact on the vehicle, surround sounds, approach identification information for enabling an approach of an emergency vehicle to be identified, and the like as surrounding environment information of the vehicle. A vehicle control processing unit 61 recognizes predetermined external stimulation information included in the surrounding environment information, for example, a specific signal or a movement with a specific pattern, a contact to a specific part, and a specific sound. The vehicle control processing unit 61 uses the recognized predetermined external stimulation information as a vehicle brake instruction from outside and starts a vehicle braking sequence to stop the vehicle. Even if an abnormality occurs in an automatically driven vehicle and the like, vehicle brake control is easily applied to a traveling vehicle from outside.

Smart wearable devices and methods for automatically configuring capabilities with biology and environment capture sensors

Abstract: A smart wearable device with sensors that can acquire biological input about the user as well as environmental data is presented. Theses sensors can acquire sensor input which causes the smart wearable device to identify a required task to perform. If the smart wearable device determines that the task can only be performed properly by acquiring new capabilities, the smart wearable device can automatically acquire the necessary capabilities from various data sources and configure itself to perform the task properly.

Optical device, head mounted display, assembling method for the same, holographic diffraction grating, display device, and alignment device

Abstract: In some embodiments, a head mounted display is provided, comprising a light guide, at least first and second diffraction gratings, alignment marks A1 and B1 associated with the first diffraction grating, and alignment marks A2 and B2 associated with the second diffraction grating. The first diffraction grating may be disposed on a first surface of the light guide, the first surface facing away from a light input. The second diffraction grating may be disposed on a second surface of the light guide, the second surface facing towards the light input. The second diffraction grating may be adapted to diffract at least some of the light input into the light guide, and the first diffraction grating may be adapted to diffract at least some light transmitted through the light guide back into the light guide.

System and method to provide driving assistance

Abstract: Various aspects of a system and method to provide driving assistance are disclosed herein. The system comprises one or more circuits in an electronic control unit used in a vehicle configured to detect one or more actions associated with the vehicle based on one or more signals generated by one or more sensors embedded in the vehicle. The electronic control unit may be further configured to control the activation of an imaging device located on the vehicle at an opposite side of a driver of the vehicle, based on the detected one or more actions. The imaging device is operable to capture one or more images in a forward-facing direction of the vehicle.

Light guide unit, image display device, and display apparatus

Abstract: An image display device includes an image forming unit and a light guide unit. The light guide unit includes a light guide plate and a housing member. The light guide plate is configured to guide light of an image formed in the image forming unit. The housing member is configured to house the light guide plate. The housing member includes a first cover member, a second cover member, and an attachment member. The second cover member is attached to the first cover member. The attachment member is configured to attach the light guide plate to a housing unit that is surrounded by the first cover member and the second cover member.

Touch screen touch force measurement based on finger deformation speed

Abstract: A device includes a capacitive sampling unit that measures, at multiple instances of time during a touch interval, capacitance values of a capacitive touch screen display associated with an area in contact with a finger touching the capacitive touch screen display. The device further includes a touch deformation area unit that determines, at the multiple instances of time, a size of the area upon the capacitive touch screen display in contact with the finger touching the capacitive touch screen display. The device also includes a touch force estimation unit that determines a rate of change in the size of the area in contact with the finger, and estimates a touching force of the finger touching the capacitive touch screen display based on the determined rate of change in the size of the area.

Semiconductor device, method of manufacturing the same, in-millimeter-wave dielectric transmission device, method of manufacturing the same, and in-millimeter-wave dielectric transmission system

Abstract: A millimeter-wave dielectric transmission device. The millimeter-wave dielectric transmission device includes a semiconductor chip provided on one interposer substrate and capable of millimeter-wave dielectric transmission, an antenna structure connected to the semiconductor chip, two semiconductor packages including a molded resin configured to cover the semiconductor chip and the antenna structure, and a dielectric transmission path provided between the two semiconductor packages to transmit a millimeter wave signal. The semiconductor packages are mounted such that the antenna structures thereof are arranged with the dielectric transmission path interposed therebetween.

Affinity CNN: Learning Pixel-Centric Pairwise Relations for Figure/Ground Embedding

Abstract: Spectral embedding provides a framework for solving perceptual organization problems, including image segmentation and figure/ground organization. From an affinity matrix describing pairwise relationships between pixels, it clusters pixels into regions, and, using a complex-valued extension, orders pixels according to layer. We train a convolutional neural network (CNN) to directly predict the pairwise relationships that define this affinity matrix. Spectral embedding then resolves these predictions into a globally-consistent segmentation and figure/ground organization of the scene. Experiments demonstrate significant benefit to this direct coupling compared to prior works which use explicit intermediate stages, such as edge detection, on the pathway from image to affinities. Our results suggest spectral embedding as a powerful alternative to the conditional random field (CRF)-based globalization schemes typically coupled to deep neural networks.