[go: up one dir, main page]

WO2023191768A1 - Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout - Google Patents

Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout Download PDF

Info

Publication number
WO2023191768A1
WO2023191768A1 PCT/US2022/022228 US2022022228W WO2023191768A1 WO 2023191768 A1 WO2023191768 A1 WO 2023191768A1 US 2022022228 W US2022022228 W US 2022022228W WO 2023191768 A1 WO2023191768 A1 WO 2023191768A1
Authority
WO
WIPO (PCT)
Prior art keywords
video data
video
analysis
data
compression
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2022/022228
Other languages
English (en)
Inventor
Radmilo Bozinovic
Isaak E. VAN KEMPEN
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Eyes Technology Inc
Original Assignee
Eyes Technology Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Eyes Technology Inc filed Critical Eyes Technology Inc
Priority to PCT/US2022/022228 priority Critical patent/WO2023191768A1/fr
Priority to JP2024557903A priority patent/JP2025510376A/ja
Publication of WO2023191768A1 publication Critical patent/WO2023191768A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/168—Segmentation; Edge detection involving transform domain methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/62—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding by frequency transforming in three dimensions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • G06T2207/20064—Wavelet transform [DWT]

Definitions

  • Vast amounts of video data are produced daily from a broad range of devices on an unprecedented scale, regardless of whether interpretation of this information is required in real time. Not only is the volume of data rapidly growing, but also the variety (range of data types, resolutions, sources) and velocity (rate of capture and transmission), leading to the well-known “3 Vs of Big Data”, originally formulated by Gartner. While some level of real- time automated video analysis has become widespread in order to facilitate useful processing of this data, there still remains a huge gap and thus growth potential in surveillance, automotive, robotics, medical/wellness, etc.
  • edge computing effectively implements the “edge computing” paradigm, thus justifying its succinct summary from an authoritative survey paper, whereby “the rationale of edge computing is that computing should happen at the proximity of data sources” (Shi et al.: Edge Computing Vision and Challenges, IEEE IoT Journal, Vol.3, No.5, Oct.2016) and fulfilling one its key benefits on “data abstraction”, whereby “data should be preprocessed at the gateway level, such as noise/low-quality removal, event detection, and privacy protection, and so on”.
  • the present invention describes a method to optimally integrate two hitherto separate and sequentially performed operations: compression and low-level analysis of video. This method is embedded in a camera subsystem, designed to facilitate certain real-time semantic human activity recognition and communication with remote portable device users.
  • Events of Interest could alternatively involve the detection of: intrusion/trespassing (unauthorized or otherwise suspicious personnel in restricted areas); tailgating (illegal follow-through at controlled access points); suspicious objects (unattended items deposited in a sensitive area beyond a predefined limit); loitering (persons identified in a controlled area for an extended time period); possibly other select predefined anomalous activity (e.g. violent physical contact, persons lying on the ground).
  • the present invention consequently embodies aspects of the edge computing paradigm, along with elements of edge analytics and cloudlet functionality (e.g., see M. Satyanarayanan: The Emergence of Edge Computing, IEEE Computer, 2017) The present invention introduces such refinements.
  • the present invention has several aspects or facets that can be used independently, although they are preferably employed together to optimize their benefits. All of the foregoing operational principles and advantages of the present invention will be more fully appreciated upon consideration of the following detailed description, with reference to the drawings.
  • SUMMARY OF THE INVENTION This invention presents a method and apparatus that integrates video compression and low-level video analysis to determine specific behavior of people. Such apparatus can operate autonomously in a broad range of applications that require real-time automated analysis of human activity backed by original footage and coupled with considerations of privacy and security.
  • AAL Ambient Assisted Living
  • AAL Ambient Assisted Living
  • AAA authentication, authorization and accounting
  • This embodiment selects and employs modules to accommodate the applicant’s basic data flow, which includes: user registration, authentication, authorization and accounting, facilitates secure uploading, management and archiving of user data files and performs Big-Data processing on uploaded data files per business requirements of the remotely connected users.
  • This invention presents an apparatus and a method to process video data comprising: (a) using a camera with a computing device with a memory storage and a power source to capture and to store the video data; (b) simultaneously compressing and analyzing said video data using Low Level Analysis of contours of static and moving objects within the video data; without limitation: said compression can be wavelet-based; said analysis can use or employ contour extraction and edge detection; and said compression can integrate or employ said analysis within the workflow of said compression.
  • the camera can be wirelessly connected to a base station and can communicate through 2-way full duplex connection; the camera can emit and stream outline video data and ternary gray-level images, flat backgrounds, outlines of static objects or outlines of moving objects; a learning module can be used or trained to recognize units and events and trends of said units and said events over time; the learning module can be used or trained to recognize long-term changes in activities of daily living, gait, stoop, ambulation and get-up time; or long-term changes in audio data and data from binary, discrete or continuous-valued sensors; or long-term changes in ambient environment, lights, doors, windows and room furnishings.
  • the analysis of the video data can compare real-time semantic human activity within the video data, whereby the method can identify the specific human activity within the video data.
  • the video data can be a video sequence of two-dimensional NxK arrays of numerical values; and using low level analysis to produce salient outline features of moving and static scene objects and humans, the method can provide for automated scene analysis of said video sequence; the method can also use or employ real-time compression and automated visual analysis of a source video signal of two-dimensional arrays of numerical values.
  • the applicant’s network server application can have the following main functions: (1) Interface with AAA server to register and manage users and base stations (clients); (2) Upload, store and manage data files from registered base stations; (3) Enable registered users to log in and view data files / statistics / metrics from mapped base stations; (4) Provide following administrative functions: Add new users and register them with AAA server; Add new clients (base stations) and register them with AAA server; Provide method to map users to base stations; (one-to-one, on-to-many and many-to-many); Provide ability to view upload directories; Provide logging for all server actions; Enable defining retention time for uploaded files.
  • Fig.1 shows a client subsystem, consisting of a camera module which captures outline or full video data and is connected to a base station;
  • Fig.2 shows one preferred embodiment of the data sequence;
  • Fig.3 shows a chart with an Initial Deployment Proposed Technology Stack;
  • Fig.4 shows a Network Server Deployment Diagram; Terms and Acronyms: AWS – Amazon Web Services; VPC – Virtual Private Cloud; EC2 – Elastic Compute Cloud—AWS Server; EFS – Elastic File Storage; RDS – Relational Database Service; ELB – Elastic Load Balancer;
  • Fig.5 shows a chart with Basic Deployment Facts;
  • Fig.6 shows an overview of the camera module performing simultaneous compression and low-level video analysis by piping data from an imaging sensor first through both temporal and spatial decomposition, and subsequent processing;
  • Fig.7 shows an overview of a preferred embodiment of the invention with the camera module, base station, network server and decoder
  • the present invention describes a camera apparatus (“camera module” – CM) with the ability to emit/stream Outline Video (OV; here understood to mean a ternary gray-level image displaying: / flat backgrounds; / outlines of static objects; / outlines of moving objects; / each at a different gray level) while simultaneously recording a compressed version of Full Video (FV) in its available internal memory (ring buffer).
  • OV Outline Video
  • FV Full Video
  • the camera module can be multiple units or any audio and video input or data gathering device.
  • the standalone quality (and utility) of this embodiment stems from the Outline Video being both human perceptible and meaningful (apart from being input and first stage of possible further processing, to which it can organically blended). Additionally, the utility of having Outline Video alone made available to users is practically justified for certain applications, specifically those where considerations of privacy and security are important. Further, nothing in this description conceptually precludes the Camera Module from streaming Full Video instead of Outline Video.
  • CM Camera Module
  • BS base station
  • CM -> BS the usual content is Outline Video, which is used on Base Station as a first step in a broad range of real-time semantic human activity recognition.
  • EOI Events of Interest
  • CM ⁇ ⁇ BS connection can be wireless (Wi Fi LAN) or through a wired or other wireless communication connection.
  • Third Embodiment An overall Ambient Assisted Living (AAS) system intended to facilitate Aging in Place (AiP) comprising, in addition to the client subsystem integrating CM and BS as described above, a network server (NS) and at least one or multiple mobile viewers (MV).
  • AAS Ambient Assisted Living
  • the network server provides basic authentication, authorization and accounting (AAA) functions for registered users and Base Stations, hosts data analysis modules and presents content to these users.
  • AAA authentication, authorization and accounting
  • Mobile viewers are applications hosted on remotely connected (broadband or cellular network) mobile devices, available to registered users (e.g., family members or healthcare providers), acting in the capacity of caregivers within a broader system that enables safe Aging in Place of elderly subjects monitored.
  • Imaging sensor generating in sequence 2D raster images (color or grayscale) of some fixed frequency (in practice, no less than 10-12 fps)
  • Image processor functionally part of a camera module and physically included in the camera housing, in one embodiment implemented as a general-purpose CPU/GPU with memory and appropriate software loaded) that cumulatively performs the “t+2D” spatio- temporal decomposition (alternatively: DWT-based dyadic decomposition/compression) by performing separately the following two decompositions in sequence: temporal and spatial.
  • Such methods are generally known and available to one skilled in the art (e.g. as disclosed in: Jens-Rainer Ohm, Mihaela van der Schaar, John W.
  • L lowpass
  • H highpass
  • a Haar basis can be used for wavelet decomposition along the temporal axis; however, other wavelet schemes are also possible for this unit.
  • a T-level decomposition would require a temporal block of length 2 T frames. In practice, the value of T should not be high (e.g., 1-3).
  • this temporal hierarchical decomposition can be further leveraged, by performing optional analogous thresholding operations to earlier H components, thus providing outlines of faster moving objects/parts, and thereby a measure of temporal flexibility and scalability.
  • Unit that further performs, using the said L* array/image as input, a standard multi-level spatial DWT-based dyadic decomposition/compression, into 4 bands per spatial level; this creates on each level the standard LL, LH, HL and HH bands (i.e., 2D arrays), the last 3 considered “high-frequency”, and the first “low-frequency” bands; on the final (coarsest) level, consistent with 220, these bands are denoted LL*, LH*, HL* and HH*.
  • the number of spatial levels - while naturally also dependent on the initial resolution of the image - should be kept such that it enables retaining the basic features of the objects being sought (e.g., basic outlines and poses of the human figures under observation).
  • this spatial hierarchical decomposition can be further leveraged, by performing optional analogous thresholding operations to earlier H components, thus providing coarser outlines of static objects, and thereby a measure of spatial flexibility and scalability.
  • [300] Conceptually separate integration module, where the 3 high-frequency bands are first (optionally) subjected to a certain thresholding operation; it then connects (superimposes) these 3 high-frequency bands into a single outline image, leaving a binary image of static outlines to be later transmitted further to base station 500; finally, it further superimposes onto this the said H* array, adjusting for common spatial resolution as needed.
  • the resulting outline view 2D array will (in one embodiment tracked here) then contain ternary pixels, displaying (on a black background - assumed without loss of generality) concurrently outlines of static objects (of one gray-white level) and outlines of moving objects (of another gray-white level).
  • the outline view pixels need not be restricted to 3 values, with the added range of “gray” values indicating different motion speed levels
  • Circular (ring) memory buffer where a continuously updated/overwritten compressed full video (of predefined duration ⁇ t) is stored, by using any standard techniques for quantization and entropy coding of the temporal and spatial L and H bands, hierarchically generated in the units 200-240 described above
  • Base station that receives (via wired or wireless communication - in one embodiment, through a WiFi port) a sequence of outline “ternary” images and performs ongoing automated vision analysis for the purpose of generative decision making, i.e.
  • this module in a preferred embodiment is data abstracted from video input, its receiving and integrating functionality is not restricted to that domain, and generalizing it to include other data input (including audio and miscellaneous binary, discrete and continuous-valued sensors - wearable, instrument monitoring or ambient) should be straightforward to one skilled in the art.
  • higher-level video analysis modules are based on the widely accepted (e.g., see 2011 Springer book “Visual Analysis of Humans”) classification of the four main types of activities, specifically: kinesics (movement of people or objects), proxemics (the proximity of people with each other or objects), haptics (people–object contact) and chronemics (change with time).
  • HMM Hidden Markov Model network
  • FSM FSM
  • CRF quantitative inference network
  • a state machine Starting with an initial/prior state of scene without humans present and with known inventory and layout of objects present, constantly scan consecutive frames for human presence.
  • Chronemic change of state (CCS) occurs with appearance of humans (1 or more) - either from a predefined physical entry point of known position (e.g., door) or from predefined geometric entry into the frame (e.g., left/right mid- level).
  • CCS Chronemic change of state
  • each human is tracked separately, with their own state machine (though interactions are possible, see later), until a reverse CCS occurs (i.e., human exits the frame).
  • KCS Kinesic change of state
  • PCS Proxemic change of state
  • HCS Haptic change of state
  • CCS and KCS detection begin from outline view, indicating outlines of both static and moving objects (the latter possibly with a speed indicator, as defined for module 300).
  • moving object outlines are piped through a standard Histogram-of-Gradients (HOG) calculation and SVM-like (Support Vector Machine) optimal classification, although other implementations are possible as part of this invention. Results get combined as needed with static outlines.
  • HOG Histogram-of-Gradients
  • tracking is based on a recursive Kalman-filter approach of types well known to one skilled in the art, simplified by close temporal frame proximity (high frame rate) and multi-scale representation.
  • PCS detection is based on combining results from moving and static outlines.
  • HCS detection might further employ basic human extremity detection and tracking, based on the same principles used for overall human outlines.
  • EOIs can be programmed to be triggered at any node of the state machine 510 (e.g., falling, running, tailgating, leaving a briefcase/backpack, telephoning, sitting, waving, human group size over a limit) under defined conditions.
  • Client portable device or mobile viewer (possibly to be preceded in the chain by a server entity 550, with certain user account/credential/distribution capabilities typical for servers) that can have the outline video (and full video, following a possible EOI detection) sent to it in real time.
  • This unit is equipped with some suitable dashboard-based user interface (UI), but at a minimum should allow for the said viewing of outline video, and be equipped with an instance of the decoder 900 to facilitate real-time viewing of full video, transmitted in compressed form following an EOI.
  • UI dashboard-based user interface
  • a learning module trained to recognize (or, operate on) units, events and their trends, as defined and indexed in 500 and 700, with the following main goals: a) improving recognition of EOI (on the supervised basis of possible false positives, etc.); b) determining long-term changes in activities of daily living (ADL), as well as gait, stoop, ambulation etc.
  • a learning module providing “big data analysis”, i.e.
  • This unit can be detached physically, and the point in time when such decompression occurs is independent of the real-time workflow of input signals described above.
  • it can be attached to or operating as part of mobile viewer 600, for the purpose of viewing full video (FV) footage, either in real time (as might come through BS 500 following an EOI) or subsequently accessing ring buffer 400 with its stored footage directly preceding any such EOI.
  • FV full video
  • the present invention is not limited to a fully specific codec (compression-decompression) method, and might encompass all such methods from the class of those DWT-based subband approaches, including both lossless and lorry variants thereof (the JPEG2000 standard most prominent among them);
  • compression-decompression compression-decompression
  • This invention refers to computing programs, applications or software, which are all synonymous and are used interchangeably. This invention can be applied to any computing device that is connected to a communication network or the Internet via wired or wireless connection.
  • the embodiments of the invention may be implemented by a processor-based computer system.
  • the system includes a database for receiving and storing information from users and application software for users and displaying feedback information.
  • a computer system operates to execute the functionality for server component; a computer system includes a processor and memory and disk storage.
  • Memory stores computer program instructions and data.
  • Processor executes the program instructions or software, and processes the data stored in memory.
  • Disk storage stores data to be transferred to and from memory.
  • disk storage can be used to store data that is typically stored in the database. All these elements are interconnected by one or more buses, which allow data to be intercommunicated between the elements.
  • memory is accessible by processor over a bus and includes: an operating system, a program partition and a data partition.
  • the program partition stores and allows execution by processor of program instructions that implement the functions of each respective system described herein.
  • the data partition is accessible by processor and stores data used during the execution of program instructions.
  • Memory, flash and disk are machine readable mediums and could include any medium capable of storing instructions adapted to be executed by a processor.
  • Some examples of such media include, but are not limited to, read-only memory (ROM), random- access memory (RAM), programmable ROM, erasable programmable ROM, electronically erasable programmable ROM, dynamic RAM, flash memory, magnetic disk (e.g., disk and hard drive), optical disk (e.g., CD-ROM), optical fiber, electrical signals, light wave signals, radio-frequency (RF) signals and any other device or signal that can store digital information.
  • the instructions are stored on the medium in a compressed and/or encrypted format.
  • a computer system also includes a network interface.
  • Network interface may be any suitable means for controlling communication signals between network devices using a desired set of communications protocols, services and operating procedures. Communication protocols are layered, which is also referred to as a protocol stack, as represented by operating system, a CBE-communication layer, and a Transport Control Protocol/Internet Protocol (TCP/IP) layer.
  • TCP/IP Transport Control Protocol/Internet Protocol
  • Network interface may also include connectors for connecting interface with a suitable communications medium.
  • network interface may receive communication signals over any suitable medium such as twisted-pair wire, co-axial cable, fiber optics, radio-frequencies, and so forth.
  • a typical computer system includes a processor, a memory, disk or flash storage, a network interface, and a protocol stack having a CBE-communication layer and a TCP/IP layer. These elements operate in a manner similar to the corresponding elements for computer system.
  • the term “plurality” is defined as: two or more than two.
  • the term “another” is defined as: at least a second or more.
  • the terms “including” and/or “having” are defined as comprising (i.e., open language).
  • the term “coupled” is defined as connected, although not necessarily directly, and not necessarily mechanically. Any element in a claim that does not explicitly state “means for” performing a specific function, or “step for” performing a specific function, is not be interpreted as a “means” or “step” clause as specified in 35 U.S.C. Sec.112, Paragraph 6. The use of “step of” in the claims is not intended to invoke the provisions of 35 U.S.C. Sec.112, Paragraph 6.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Psychiatry (AREA)
  • Social Psychology (AREA)
  • Human Computer Interaction (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

Procédé et appareil d'analyse de données vidéo comprenant : l'utilisation d'une caméra avec un dispositif informatique avec un stockage de mémoire et une source d'alimentation pour capturer et stocker les données vidéo ; la compression et l'analyse simultanée desdites données vidéo à l'aide d'une analyse de faible niveau de contours d'objets statiques et mobiles à l'intérieur des données vidéo ; ladite compression peut être une décomposition basée sur des ondelettes ; et ladite analyse des données vidéo compare une activité humaine sémantique en temps réel dans les données vidéo, moyennant quoi le procédé identifie l'activité humaine spécifique dans les données vidéo ; la caméra peut être connectée sans fil à une station de base et communique par l'intermédiaire d'une connexion en duplex intégral à 2 voies.
PCT/US2022/022228 2022-03-28 2022-03-28 Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout Ceased WO2023191768A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/US2022/022228 WO2023191768A1 (fr) 2022-03-28 2022-03-28 Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout
JP2024557903A JP2025510376A (ja) 2022-03-28 2022-03-28 リアルタイム統合「ｔ＋２Ｄ」ＤＷＴビデオ圧縮およびエンド・ツー・エンドＡＡＬシステムにオプションで組み込まれたインテリジェントカメラ内での低レベルビデオ分析のための方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2022/022228 WO2023191768A1 (fr) 2022-03-28 2022-03-28 Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout

Publications (1)

Publication Number Publication Date
WO2023191768A1 true WO2023191768A1 (fr) 2023-10-05

Family

ID=88202881

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2022/022228 Ceased WO2023191768A1 (fr) 2022-03-28 2022-03-28 Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout

Country Status (2)

Country Link
JP (1) JP2025510376A (fr)
WO (1) WO2023191768A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070013776A1 (en) * 2001-11-15 2007-01-18 Objectvideo, Inc. Video surveillance system employing video primitives
US20080310742A1 (en) * 2007-06-15 2008-12-18 Physical Optics Corporation Apparatus and method employing pre-ATR-based real-time compression and video frame segmentation
US20180204111A1 (en) * 2013-02-28 2018-07-19 Z Advanced Computing, Inc. System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2003056518A1 (fr) * 2002-01-04 2003-07-10 Zeugma Technologies Inc. Compression d'image
JP2006246200A (ja) * 2005-03-04 2006-09-14 Fuji Photo Film Co Ltd 画像処理装置および画像処理方法
CN102004922B (zh) * 2010-12-01 2012-12-26 南京大学 基于骨架特征的高分辨率遥感影像飞机提取方法
US8976269B2 (en) * 2012-06-22 2015-03-10 California Institute Of Technology Compressive sensing based bio-inspired shape feature detection CMOS imager
CN113053496B (zh) * 2021-03-19 2023-08-29 深圳高性能医疗器械国家研究院有限公司 一种用于医学图像低剂量估计的深度学习方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070013776A1 (en) * 2001-11-15 2007-01-18 Objectvideo, Inc. Video surveillance system employing video primitives
US20080310742A1 (en) * 2007-06-15 2008-12-18 Physical Optics Corporation Apparatus and method employing pre-ATR-based real-time compression and video frame segmentation
US20180204111A1 (en) * 2013-02-28 2018-07-19 Z Advanced Computing, Inc. System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform

Also Published As

Publication number Publication date
JP2025510376A (ja) 2025-04-14

Similar Documents

Publication Publication Date Title
KR102535098B1 (ko) 이미지 프로세싱 및 비디오 압축 방법
US7617167B2 (en) Machine vision system for enterprise management
US11968412B1 (en) Bandwidth estimation for video streams
US9106977B2 (en) Object archival systems and methods
US11321568B1 (en) Method for real-time integrated t+2D DWT video compression and low-level video analysis within an intelligent camera optionally embedded in an end-to-end AAL system
US20070122000A1 (en) Detection of stationary objects in video
Abdullah et al. Traffic monitoring using video analytics in clouds
US20240037778A1 (en) Video analytics accuracy using transfer learning
US20130279598A1 (en) Method and Apparatus For Video Compression of Stationary Scenes
FR2851397A1 (fr) Procede et dispositif d'analyse de sequences video dans un reseau de communication
WO2013131189A1 (fr) Analytique vidéo à base de nuage avec post-traitement au niveau de l'extrémité source.
Poularakis et al. Efficient motion estimation methods for fast recognition of activities of daily living
Elgamal et al. Sieve: Semantically encoded video analytics on edge and cloud
CN101790717A (zh) 用于企业管理的机器视觉系统
KR20230145920A (ko) 듀얼 이미지 모델을 사용하는 이미지 코딩을 위한 시스템 및 방법
WO2023191768A1 (fr) Procédé de compression vidéo dwt « t+2d » intégrée en temps réel et d'analyse vidéo de bas niveau à l'intérieur d'une caméra intelligente intégrée dans un système aal de bout en bout
US11943300B2 (en) Software-defined sensing
Paul et al. Why is the video analytics accuracy fluctuating, and what can we do about it?
Ahmed et al. Efficient and effective automated surveillance agents using kernel tricks
Constantinou et al. A crowd-based image learning framework using edge computing for smart city applications
Fang et al. PIB: Prioritized information bottleneck framework for collaborative edge video analytics
Ravindran et al. A fog-assisted framework for intelligent video preprocessing in cloud-based video surveillance as a service
Ahmed et al. Real-time intruder detection in surveillance networks using adaptive kernel methods
Kokadwar et al. Camera based smart surveillance system-literature survey
Blasch et al. Multi-modal video fusion for context-aided tracking

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22935959

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2024557903

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22935959

Country of ref document: EP

Kind code of ref document: A1