Download
Kinetics-I3D, developed by Google DeepMind, provides trained models and implementation code for the Inflated 3D ConvNet (I3D) architecture introduced in the paper “Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset” (CVPR 2017). The I3D model extends the 2D convolutional structure of Inception-v1 into 3D, allowing it to capture spatial and temporal information from videos for action recognition. This repository includes pretrained I3D models on the Kinetics dataset, with both RGB and optical flow input streams. The models have achieved state-of-the-art results on benchmark datasets such as UCF101 and HMDB51, and also won first place in the CVPR 2017 Charades Challenge. The project provides TensorFlow and Sonnet-based implementations, pretrained checkpoints, and example scripts for evaluating or fine-tuning models. It also offers sample data, including preprocessed video frames and optical flow arrays, to demonstrate how to run inference and visualize outputs.
Features
- Pretrained I3D (Inflated 3D ConvNet) models for video action recognition
- Two-stream (RGB + Flow) architecture for spatial-temporal learning
- TensorFlow + Sonnet implementation with pretrained checkpoints
- Demonstration scripts for evaluation and testing on sample data
- Compatible with Kinetics-400 and Kinetics-600 datasets
- State-of-the-art benchmark performance on UCF101 and HMDB51
Project Samples
