High frame rate action classification and state tracking in competitive sports using SlowFast dual-stream mesh structure

Mingdong Liang, Shuran Wang · 2026

Read High frame rate action classification and state tracking in competitive sports using SlowFast dual-stream mesh structure by Mingdong Liang, Shuran Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13454603.txt.

Description

Mingdong Liang and Shuran Wang proposed a SlowFast dual-stream mesh method in 2026 for high frame rate action classification in table tennis. Rapid movements and frequent state changes challenge models that must capture both meaning and motion information at high frame rates. Current methods separate semantic and dynamic features, weakening classification and state modeling.

Video frames were divided into spatial grids with fixed temporal indexing. The Slow branch extracted semantic temporal features at low sampling rate, while the Fast branch extracted fine-grained dynamic features at high sampling rate. A grid fusion module aligned spatial features while preserving temporal indices and channel attention, and a gated recurrent unit modeled state sequences. The unified approach learns motion classification and phase-state evolution together. Sports analysts gain a method for fast, continuous action understanding.

Catalog record

CREATORMingdong Liang, Shuran Wang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE151 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13454603

Browse the full library · About the Tower of Babel