---
datasets:
- moving-mnist
- taxibj
- kth
- human3.6m
license: mit
pipeline_tag: image-to-video
tags:
- computer-vision
- video-prediction
- spatiotemporal-prediction
- pytorch
paper:
- https://huggingface.co/papers/2602.20537
---

# PFGNet: A Fully Convolutional Frequency-Guided Peripheral Gating Network for Efficient Spatiotemporal Predictive Learning

PFGNet is a fully convolutional framework for efficient spatiotemporal predictive learning (STPL), presented at CVPR 2026. It aims to forecast future frames from past observations by dynamically modulating receptive fields through pixel-wise frequency-guided gating. 

Inspired by biological center-surround organization, the core Peripheral Frequency Gating (PFG) block extracts localized spectral cues to adaptively fuse multi-scale large-kernel peripheral responses with learnable center suppression, forming spatially adaptive band-pass filters.

**Resources:**
- **Paper:** [PFGNet: A Fully Convolutional Frequency-Guided Peripheral Gating Network for Efficient Spatiotemporal Predictive Learning](https://huggingface.co/papers/2602.20537)
- **Code:** [Official GitHub Repository](https://github.com/fhjdqaq/PFGNet)
- **Project Page:** [kaimaoge.github.io](https://kaimaoge.github.io)

## Available checkpoints

This repository provides dataset-specific trained checkpoints of PFGNet on multiple benchmarks:

| Dataset | Checkpoint |
|---|---|
| Moving MNIST | `pfg_mmnist.ckpt` |
| Moving Fashion MNIST | `pfg_mfmnist.ckpt` |
| TaxiBJ | `pfg_taxibj.ckpt` |
| KTH (10→20) | `pfg_kth20.ckpt` |
| KTH (10→40) | `pfg_kth40.ckpt` |
| Human3.6M | `pfg_human.ckpt` |

## Usage

PFGNet directly inherits the codebase and dependencies of [OpenSTL](https://github.com/chengtan9907/OpenSTL). Please refer to the official repository for detailed environment setup and data preparation instructions.

### Training (Moving MNIST example)
From the repository root, run:
```bash
python tools/train.py -d mmnist -c configs/mmnist/PFG.py --ex_name mmnist_pfg --test
```

### Testing (Moving MNIST example)
From the repository root, run:
```bash
python tools/test.py -d mmnist -c configs/mmnist/PFG.py --ex_name mmnist_pfg --test
```

## Citation

If you find this work helpful, please consider citing:

```bibtex
@misc{cai2026pfgnetfullyconvolutionalfrequencyguided,
      title={PFGNet: A Fully Convolutional Frequency-Guided Peripheral Gating Network for Efficient Spatiotemporal Predictive Learning}, 
      author={Xinyong Cai and Changbin Sun and Yong Wang and Hongyu Yang and Yuankai Wu},
      year={2026},
      eprint={2602.20537},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}
```