Matches in SemOpenAlex for { <https://semopenalex.org/work/W3106981029> ?p ?o ?g. }
- W3106981029 abstract "Modern methods for counting people in crowded scenes rely on deep networks to estimate people densities in individual images. As such, only very few take advantage of temporal consistency in video sequences, and those that do only impose weak smoothness constraints across consecutive frames. In this paper, we advocate estimating people flows across image locations between consecutive images and inferring the people densities from these flows instead of directly regressing them. This enables us to impose much stronger constraints encoding the conservation of the number of people. As a result, it significantly boosts performance without requiring a more complex architecture. Furthermore, it allows us to exploit the correlation between people flow and optical flow to further improve the results. We also show that leveraging people conservation constraints in both a spatial and temporal manner makes it possible to train a deep crowd counting model in an active learning setting with much fewer annotations. This significantly reduces the annotation cost while still leading to similar performance to the full supervision case." @default.
- W3106981029 created "2020-12-07" @default.
- W3106981029 creator A5038674741 @default.
- W3106981029 creator A5049300388 @default.
- W3106981029 creator A5063856911 @default.
- W3106981029 date "2020-12-01" @default.
- W3106981029 modified "2023-09-26" @default.
- W3106981029 title "Counting People by Estimating People Flows" @default.
- W3106981029 cites W1485009520 @default.
- W3106981029 cites W1607775576 @default.
- W3106981029 cites W1840672702 @default.
- W3106981029 cites W1903029394 @default.
- W3106981029 cites W1910776219 @default.
- W3106981029 cites W1987389530 @default.
- W3106981029 cites W2016135469 @default.
- W3106981029 cites W2017849585 @default.
- W3106981029 cites W2064675550 @default.
- W3106981029 cites W2083049794 @default.
- W3106981029 cites W2099471712 @default.
- W3106981029 cites W2111644456 @default.
- W3106981029 cites W2112547580 @default.
- W3106981029 cites W2113929305 @default.
- W3106981029 cites W2115734113 @default.
- W3106981029 cites W2115739848 @default.
- W3106981029 cites W2123175289 @default.
- W3106981029 cites W2127084114 @default.
- W3106981029 cites W2129566274 @default.
- W3106981029 cites W2133601766 @default.
- W3106981029 cites W2145983039 @default.
- W3106981029 cites W2154434560 @default.
- W3106981029 cites W2164990725 @default.
- W3106981029 cites W2171243491 @default.
- W3106981029 cites W2400262625 @default.
- W3106981029 cites W2463631526 @default.
- W3106981029 cites W2519281173 @default.
- W3106981029 cites W2519354012 @default.
- W3106981029 cites W2520826941 @default.
- W3106981029 cites W2532516272 @default.
- W3106981029 cites W2541389513 @default.
- W3106981029 cites W2741077351 @default.
- W3106981029 cites W2753789229 @default.
- W3106981029 cites W2798489385 @default.
- W3106981029 cites W2798490576 @default.
- W3106981029 cites W2798781811 @default.
- W3106981029 cites W2798820905 @default.
- W3106981029 cites W2808519136 @default.
- W3106981029 cites W2884960332 @default.
- W3106981029 cites W2886443245 @default.
- W3106981029 cites W2895051362 @default.
- W3106981029 cites W2903922555 @default.
- W3106981029 cites W2914913933 @default.
- W3106981029 cites W2948513880 @default.
- W3106981029 cites W2949333977 @default.
- W3106981029 cites W2955171701 @default.
- W3106981029 cites W2962854645 @default.
- W3106981029 cites W2963035940 @default.
- W3106981029 cites W2963049075 @default.
- W3106981029 cites W2963231953 @default.
- W3106981029 cites W2963693541 @default.
- W3106981029 cites W2963782415 @default.
- W3106981029 cites W2963838390 @default.
- W3106981029 cites W2963866368 @default.
- W3106981029 cites W2963893037 @default.
- W3106981029 cites W2964018834 @default.
- W3106981029 cites W2964059111 @default.
- W3106981029 cites W2964203052 @default.
- W3106981029 cites W2964209782 @default.
- W3106981029 cites W2964721919 @default.
- W3106981029 cites W2966271765 @default.
- W3106981029 cites W2967069910 @default.
- W3106981029 cites W2967776630 @default.
- W3106981029 cites W2975221988 @default.
- W3106981029 cites W2978859332 @default.
- W3106981029 cites W2981436300 @default.
- W3106981029 cites W2982007926 @default.
- W3106981029 cites W2982014038 @default.
- W3106981029 cites W2987988567 @default.
- W3106981029 cites W2991203386 @default.
- W3106981029 cites W2995582330 @default.
- W3106981029 cites W3004672782 @default.
- W3106981029 cites W3007175960 @default.
- W3106981029 cites W3010021361 @default.
- W3106981029 cites W3013987429 @default.
- W3106981029 cites W3027606690 @default.
- W3106981029 cites W3035193053 @default.
- W3106981029 cites W3035307763 @default.
- W3106981029 cites W3049248796 @default.
- W3106981029 cites W3082125592 @default.
- W3106981029 cites W3107554785 @default.
- W3106981029 cites W3108659450 @default.
- W3106981029 cites W3108994527 @default.
- W3106981029 cites W3109157205 @default.
- W3106981029 cites W3109242411 @default.
- W3106981029 cites W3110700482 @default.
- W3106981029 cites W603908379 @default.
- W3106981029 doi "https://doi.org/10.48550/arxiv.2012.00452" @default.
- W3106981029 hasPublicationYear "2020" @default.
- W3106981029 type Work @default.
- W3106981029 sameAs 3106981029 @default.
- W3106981029 citedByCount "0" @default.