FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2990897304> ?p ?o ?g. }

• W2990897304 abstract "We address a core problem of computer vision: Detection and description of 2D feature points for image matching. For a long time, hand-crafted designs, like the seminal SIFT algorithm, were unsurpassed in accuracy and efficiency. Recently, learned feature detectors emerged that implement detection and description using neural networks. Training these networks usually resorts to optimizing low-level matching scores, often pre-defining sets of image patches which should or should not match, or which should or should not contain key points. Unfortunately, increased accuracy for these low-level matching scores does not necessarily translate to better performance in high-level vision tasks. We propose a new training methodology which embeds the feature detector in a complete vision pipeline, and where the learnable parameters are trained in an end-to-end fashion. We overcome the discrete nature of key point selection and descriptor matching using principles from reinforcement learning. As an example, we address the task of relative pose estimation between a pair of images. We demonstrate that the accuracy of a state-of-the-art learning-based feature detector can be increased when trained for the task it is supposed to solve at test time. Our training methodology poses little restrictions on the task to learn, and works for any architecture which predicts key point heat maps, and descriptors for key point locations." @default.
• W2990897304 created "2019-12-05" @default.
• W2990897304 creator A5000173125 @default.
• W2990897304 creator A5020201032 @default.
• W2990897304 creator A5034290170 @default.
• W2990897304 creator A5056030247 @default.
• W2990897304 date "2019-12-02" @default.
• W2990897304 modified "2023-09-26" @default.
• W2990897304 title "Reinforced Feature Points: Optimizing Feature Detection and Description for a High-Level Task" @default.
• W2990897304 cites W1515851193 @default.
• W2990897304 cites W1565312575 @default.
• W2990897304 cites W1616969904 @default.
• W2990897304 cites W1677409904 @default.
• W2990897304 cites W1901447884 @default.
• W2990897304 cites W1929856797 @default.
• W2990897304 cites W1979931042 @default.
• W2990897304 cites W1985238052 @default.
• W2990897304 cites W2033819227 @default.
• W2990897304 cites W2085261163 @default.
• W2990897304 cites W2103924867 @default.
• W2990897304 cites W2105303354 @default.
• W2990897304 cites W2109635530 @default.
• W2990897304 cites W2111308925 @default.
• W2990897304 cites W2115362620 @default.
• W2990897304 cites W2117228865 @default.
• W2990897304 cites W2119717200 @default.
• W2990897304 cites W2128017662 @default.
• W2990897304 cites W2134446283 @default.
• W2990897304 cites W2141362318 @default.
• W2990897304 cites W2151103935 @default.
• W2990897304 cites W2161969291 @default.
• W2990897304 cites W2172188317 @default.
• W2990897304 cites W2200124539 @default.
• W2990897304 cites W2320444803 @default.
• W2990897304 cites W2471962767 @default.
• W2990897304 cites W2522940611 @default.
• W2990897304 cites W2556455135 @default.
• W2990897304 cites W2561074213 @default.
• W2990897304 cites W2612112834 @default.
• W2990897304 cites W2737094507 @default.
• W2990897304 cites W2737260104 @default.
• W2990897304 cites W2740418457 @default.
• W2990897304 cites W2812468425 @default.
• W2990897304 cites W2884354140 @default.
• W2990897304 cites W2892865870 @default.
• W2990897304 cites W2897085414 @default.
• W2990897304 cites W2899771611 @default.
• W2990897304 cites W2922243907 @default.
• W2990897304 cites W2951870616 @default.
• W2990897304 cites W2963537932 @default.
• W2990897304 cites W2963674285 @default.
• W2990897304 cites W2963748588 @default.
• W2990897304 cites W2963856988 @default.
• W2990897304 cites W2964121744 @default.
• W2990897304 cites W2964157791 @default.
• W2990897304 cites W2973665503 @default.
• W2990897304 cites W2982190091 @default.
• W2990897304 cites W2982446862 @default.
• W2990897304 cites W2987672160 @default.
• W2990897304 cites W3001017583 @default.
• W2990897304 cites W3009931536 @default.
• W2990897304 cites W3043075211 @default.
• W2990897304 cites W3102327032 @default.
• W2990897304 cites W3210232381 @default.
• W2990897304 cites W349087979 @default.
• W2990897304 hasPublicationYear "2019" @default.
• W2990897304 type Work @default.
• W2990897304 sameAs 2990897304 @default.
• W2990897304 citedByCount "0" @default.
• W2990897304 crossrefType "posted-content" @default.
• W2990897304 hasAuthorship W2990897304A5000173125 @default.
• W2990897304 hasAuthorship W2990897304A5020201032 @default.
• W2990897304 hasAuthorship W2990897304A5034290170 @default.
• W2990897304 hasAuthorship W2990897304A5056030247 @default.
• W2990897304 hasConcept C105795698 @default.
• W2990897304 hasConcept C115961682 @default.
• W2990897304 hasConcept C119857082 @default.
• W2990897304 hasConcept C126422989 @default.
• W2990897304 hasConcept C127413603 @default.
• W2990897304 hasConcept C138885662 @default.
• W2990897304 hasConcept C153180895 @default.
• W2990897304 hasConcept C154945302 @default.
• W2990897304 hasConcept C165064840 @default.
• W2990897304 hasConcept C199360897 @default.
• W2990897304 hasConcept C201995342 @default.
• W2990897304 hasConcept C2524010 @default.
• W2990897304 hasConcept C26517878 @default.
• W2990897304 hasConcept C2776401178 @default.
• W2990897304 hasConcept C2780451532 @default.
• W2990897304 hasConcept C28719098 @default.
• W2990897304 hasConcept C31972630 @default.
• W2990897304 hasConcept C33923547 @default.
• W2990897304 hasConcept C38652104 @default.
• W2990897304 hasConcept C38785706 @default.
• W2990897304 hasConcept C41008148 @default.
• W2990897304 hasConcept C41895202 @default.
• W2990897304 hasConcept C43521106 @default.
• W2990897304 hasConcept C61265191 @default.
• W2990897304 hasConcept C76155785 @default.
• W2990897304 hasConcept C9417928 @default.

• next