Papers › Adaptive frame selection in two dimensional convolutional neural network action recognition

Adaptive frame selection in two dimensional convolutional neural network action recognition

28 Dec 2022Conference 2022 12archive 2025-07-28

Alireza Rahnama, Alireza Esfahani, Azadeh Mansouri

We presented a technique in this research for dynamic frame selection to achieve robust features. This situation results in less redundancy and useful input for the network. Because it uses fewer processing resources and offers adequate accuracy, the suggested technique is appropriate for real-time applications. The network becomes more efficient and maintains adequate accuracy when informative frames are chosen and computation is minimized. The framework is tested on UCFIOI as one of the large and realistic datasets. The experiments show acceptable results employing both Resnet-50 and Mobilenet pre-trained features.

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Results from the paper archive 2025-07-28

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections