DISTILLATION OF NEURAL NETWORK MODELS FOR DETECTION AND DESCRIPTION OF IMAGE KEY POINTS
Annotation
Subject of Research. Image matching and classification methods, as well as synchronous location and mapping, are widely used on embedded and mobile devices. Their most resource-intensive part is the detection and description of the image key points. In case of classical methods for detection and description of key points they can be executed in real time on mobile devices but for modern neural network methods with better quality, such approach is difficult due to trading off performance. Thus, the issue of speeding for neural network models applied for the detection and description of key points is currently topical. The subject of research is distillation as one of the methods for reducing neural network models. The aim of the study is to obtain more compact model for detection and description of key points and a description of the procedure for this model design. Method. We proposed a method for pairing the original and more compact new model for its subsequent training on the output values of the original model. In this regard, the new model is learned to reconstruct the output of the original model without using image labels. Both networks accept identical images as input. Main Results. Neural network distillation method for detection and description of key points is tested. The objective function and training parameters that provide the best results in the framework of the study are proposed. A new data set is created for testing key point detection methods, and a new quality indicator of the allocated key points and their corresponding local features is added. New model training in the way suggested with the same number of parameters, shows greater accuracy in key points compared to the original model. A new model with a significantly smaller number of parameters shows the accuracy of point matching close to the accuracy of the original model. Practical Relevance. More compact model for detection and description of image key points is created applying the proposed method. The model is applicable on embedded and mobile devices for synchronous location and mapping. Such model application can also increase the service efficiency of the image search on the server side.
Keywords
Постоянный URL
Articles in current issue
- EXPERIMENTAL METHOD FOR DETERMINATION OF SHRINKAGE DIRECTION DURING HOLOGRAPHIC RECORDING IN BAYFOL HX PHOTOPOLYMER
- AERIAL MAPPING BASED ON ARRANGEMENT OF OPTICAL ELECTRON CAMERAS
- Koreshev S.N., Starovoitov S.O., Smorodinov D.S., Frolova M.A.QUALITY ASSESSMENT OF BINARY OBJECT IMAGES RECONSTRUCTED BY COMPUTER-GENERATED HOLOGRAMS
- NONDESTRUCTIVE TESTING OF BALTIC AMBER:OPTICAL ANALYSIS OF MACRO- AND MICROSTRUCTURE
- FIBER OPTIC MEASUREMENT SYSTEM FOR DETERMINATION OF EXTENDED OBJECT POSITION AND BENDS IN 3D SPACE
- RECOVERY OF DISCRETE SPECTRA RADIATED BY SUBSTANCE IN DEEP VACUUM USING INTEGRAL APPROXIMATION ALGORITHM
- Omorov R.O.ROBUSTNESS RESEARCH OF INTERVAL DYNAMIC SYSTEMS BY ALGEBRAIC METHOD
- RESEARCH OF VISUAL SIMULTANEOUS LOCALIZATION AND MAPPING-BASED NAVIGATION SYSTEM FOR MOBILE ROBOTS
- REVIEW OF METHODS FOR SIZE AND MORPHOLOGY DETERMINATION OF VESICLES IN NIOSOME DISPERSION
- INFORMATION REPRESENTATION METHODS IN SIMPLE SEMANTIC NETWORKS
- DETERMINISTIC FINITE AUTOMATA USINGCOUNTEREXAMPLE GUIDED ABSTRACTION REFINEMENT
- DISTILLATION OF NEURAL NETWORK MODELS FOR DETECTION AND DESCRIPTION OF IMAGE KEY POINTS
- SOFTWARE PORTABILITY BASED ON RETARGETABLE RUNTIME ENVIRONMENt
- REAL TIME DETECTION AND CLASSIFICATION OF TRAFFIC SIGNS BASED ON YOLO VERSION 3 ALGORITHM (in English)
- U-NET ARCHITECTURE NEURAL NETWORK FOR LOCALIZATION OF DIGITAL IMAGES INTEGRITY VIOLATION
- DETERMINISTIC SYSTEMS WITH NATURAL QUANTIZATION
- TECHNICAL PNEUMOSYSTEM FOR DEVELOPMENT OF DEVICES WITH CERTAIN FUNCTIONAL CAPABILITIES
- STATISTICAL MODELING OF KNEE JOINT GEAR RATIOS
- COMPARISON OF BEAMFORMING ALGORITHMS FOR MICROPHONE ARRAYS IN MATLAB
- QR CODES WITH ANIMATION FOR DIGITAL PASSES
- PHOTOACTINIC IRRADIATION EFFECT ON REFRACTION INDICE OF ORGANIC CO-CRYSTALS BASED ON AMINOPYRIDINE SERIES