Matches in SemOpenAlex for { <https://semopenalex.org/work/W3087446705> ?p ?o ?g. }
- W3087446705 abstract "Visual localization is a crucial component in the application of mobile robot and autonomous driving. Image retrieval is an efficient and effective technique in image-based localization methods. Due to the drastic variability of environmental conditions, e.g. illumination, seasonal and weather changes, retrieval-based visual localization is severely affected and becomes a challenging problem. In this work, a general architecture is first formulated probabilistically to extract domain-invariant feature through multi-domain image translation. And then a novel gradient-weighted similarity activation mapping loss (Grad-SAM) is incorporated for finer localization with high accuracy. We also propose a new adaptive triplet loss to boost the metric learning of the embedding in a self-supervised manner. The final coarse-to-fine image retrieval pipeline is implemented as the sequential combination of models without and with Grad-SAM loss. Extensive experiments have been conducted to validate the effectiveness of the proposed approach on the CMU-Seasons dataset. The strong generalization ability of our approach is verified on RobotCar dataset using models pre-trained on urban part of CMU-Seasons dataset. Our performance is on par with or even outperforms the state-of-the-art image-based localization baselines in medium or high precision, especially under the challenging environments with illumination variance, vegetation and night-time images. Moreover, real-site experiments have been conducted to validate the efficiency and effectiveness of the coarse-to-fine strategy for localization." @default.
- W3087446705 created "2020-09-25" @default.
- W3087446705 creator A5024584263 @default.
- W3087446705 creator A5039258932 @default.
- W3087446705 creator A5054513966 @default.
- W3087446705 creator A5091161457 @default.
- W3087446705 date "2020-09-16" @default.
- W3087446705 modified "2023-10-17" @default.
- W3087446705 title "Domain-invariant Similarity Activation Map Metric Learning for Retrieval-based Long-term Visual Localization" @default.
- W3087446705 cites W1975517671 @default.
- W3087446705 cites W1989484209 @default.
- W3087446705 cites W2012592962 @default.
- W3087446705 cites W2014987111 @default.
- W3087446705 cites W2099471712 @default.
- W3087446705 cites W2110405746 @default.
- W3087446705 cites W2117154949 @default.
- W3087446705 cites W2144824356 @default.
- W3087446705 cites W2173520492 @default.
- W3087446705 cites W2284029970 @default.
- W3087446705 cites W2295107390 @default.
- W3087446705 cites W2361494151 @default.
- W3087446705 cites W2502225121 @default.
- W3087446705 cites W2522940611 @default.
- W3087446705 cites W2544587078 @default.
- W3087446705 cites W2559823555 @default.
- W3087446705 cites W2580440899 @default.
- W3087446705 cites W2589016937 @default.
- W3087446705 cites W2609368314 @default.
- W3087446705 cites W2619522967 @default.
- W3087446705 cites W2620629206 @default.
- W3087446705 cites W2679426462 @default.
- W3087446705 cites W2737075200 @default.
- W3087446705 cites W2740418457 @default.
- W3087446705 cites W2757526371 @default.
- W3087446705 cites W2765793020 @default.
- W3087446705 cites W2795414580 @default.
- W3087446705 cites W2884148180 @default.
- W3087446705 cites W2892865870 @default.
- W3087446705 cites W2902227449 @default.
- W3087446705 cites W2907017794 @default.
- W3087446705 cites W2922243907 @default.
- W3087446705 cites W2962705366 @default.
- W3087446705 cites W2962734311 @default.
- W3087446705 cites W2962793481 @default.
- W3087446705 cites W2962858109 @default.
- W3087446705 cites W2962947361 @default.
- W3087446705 cites W2962975391 @default.
- W3087446705 cites W2963026686 @default.
- W3087446705 cites W2963073614 @default.
- W3087446705 cites W2963104724 @default.
- W3087446705 cites W2963226019 @default.
- W3087446705 cites W2963344645 @default.
- W3087446705 cites W2963359064 @default.
- W3087446705 cites W2963538243 @default.
- W3087446705 cites W2963588253 @default.
- W3087446705 cites W2963713713 @default.
- W3087446705 cites W2963713828 @default.
- W3087446705 cites W2963775347 @default.
- W3087446705 cites W2963789515 @default.
- W3087446705 cites W2963890275 @default.
- W3087446705 cites W2967005398 @default.
- W3087446705 cites W2967242391 @default.
- W3087446705 cites W2968225067 @default.
- W3087446705 cites W2969050209 @default.
- W3087446705 cites W2969121634 @default.
- W3087446705 cites W2990533803 @default.
- W3087446705 cites W2993164566 @default.
- W3087446705 cites W2996967608 @default.
- W3087446705 cites W3012481664 @default.
- W3087446705 cites W3013335708 @default.
- W3087446705 cites W3038022805 @default.
- W3087446705 cites W3098873762 @default.
- W3087446705 cites W3102327032 @default.
- W3087446705 cites W3106458387 @default.
- W3087446705 cites W3159890710 @default.
- W3087446705 hasPublicationYear "2020" @default.
- W3087446705 type Work @default.
- W3087446705 sameAs 3087446705 @default.
- W3087446705 citedByCount "0" @default.
- W3087446705 crossrefType "posted-content" @default.
- W3087446705 hasAuthorship W3087446705A5024584263 @default.
- W3087446705 hasAuthorship W3087446705A5039258932 @default.
- W3087446705 hasAuthorship W3087446705A5054513966 @default.
- W3087446705 hasAuthorship W3087446705A5091161457 @default.
- W3087446705 hasConcept C103278499 @default.
- W3087446705 hasConcept C115961682 @default.
- W3087446705 hasConcept C138885662 @default.
- W3087446705 hasConcept C153180895 @default.
- W3087446705 hasConcept C154945302 @default.
- W3087446705 hasConcept C162324750 @default.
- W3087446705 hasConcept C1667742 @default.
- W3087446705 hasConcept C176217482 @default.
- W3087446705 hasConcept C190470478 @default.
- W3087446705 hasConcept C21547014 @default.
- W3087446705 hasConcept C2776401178 @default.
- W3087446705 hasConcept C31972630 @default.
- W3087446705 hasConcept C33923547 @default.
- W3087446705 hasConcept C37914503 @default.
- W3087446705 hasConcept C41008148 @default.
- W3087446705 hasConcept C41608201 @default.