Action Recognition In Videos
64 papers with code • 17 benchmarks • 17 datasets
Action Recognition in Videos is a task in computer vision and pattern recognition where the goal is to identify and categorize human actions performed in a video sequence. The task involves analyzing the spatiotemporal dynamics of the actions and mapping them to a predefined set of action classes, such as running, jumping, or swimming.
Benchmarks
These leaderboards are used to track progress in Action Recognition In Videos
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
Libraries
Use these libraries to find Action Recognition In Videos models and implementationsDatasets
UCF101
Kinetics
HMDB51
ActivityNet
Kinetics 400
NTU RGB+D
THUMOS14
Something-Something V2
Sports-1M
Kinetics-600
Latest papers
ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videos
Our framework leverages both labeled and unlabelled data to robustly learn action representations in videos, combining pseudo-labeling with contrastive learning for effective learning from both types of samples.
HaltingVT: Adaptive Token Halting Transformer for Efficient Video Recognition
Action recognition in videos poses a challenge due to its high computational cost, especially for Joint Space-Time video transformers (Joint VT).
CAST: Cross-Attention in Space and Time for Video Action Recognition
In this work, we propose a novel two-stream architecture, called Cross-Attention in Space and Time (CAST), that achieves a balanced spatio-temporal understanding of videos using only RGB input.
Actor-agnostic Multi-label Action Recognition with Multi-modal Query
Existing action recognition methods are typically actor-specific due to the intrinsic topological and apparent differences among the actors.
Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles
Modern hierarchical vision transformers have added several vision-specific components in the pursuit of supervised classification performance.
VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking
OpenGVLab/VideoMAEv2
•
•
CVPR 2023
Finally, we successfully train a video ViT model with a billion parameters, which achieves a new state-of-the-art performance on the datasets of Kinetics (90. 0% on K400 and 89. 9% on K600) and Something-Something (68. 7% on V1 and 77. 0% on V2).
Dual-path Adaptation from Image to Video Transformers
park-jungin/dualpath
•
•
CVPR 2023
In this paper, we efficiently transfer the surpassing representation power of the vision foundation models, such as ViT and Swin, for video understanding with only a few trainable parameters.
Video Action Recognition Collaborative Learning with Dynamics via PSO-ConvNet Transformer
To extend our approach to video, we integrate ConvNets with state-of-the-art temporal methods such as Transformer and Recurrent Neural Networks.
Rethinking Video ViTs: Sparse Video Tubes for Joint Image and Video Learning
daniel-code/TubeViT
•
•
CVPR 2023
We present a simple approach which can turn a ViT encoder into an efficient video model, which can seamlessly work with both image and video inputs.
Exploring Modulated Detection Transformer as a Tool for Action Recognition in Videos
We show that it is possible to use a multi-modal model to tackle a task that it was not designed for.
