♾️ CAREC

CAREC: Continual Wireless Action Recognition with Expansion-Compression Coordination

In real-world applications, user demands for new functionalities and activities constantly evolve, requiring action recognition systems to incrementally incorporate new action classes without retraining from scratch. This class-incremental learning (CIL) paradigm is essential for enabling adaptive and scalable systems that can grow over time. However, Wi-Fi-based indoor action recognition under incremental learning faces two major challenges: catastrophic forgetting of previously learned knowledge and uncontrolled model expansion as new classes are added. To address these issues, we propose CAREC, a class-incremental framework that balances dynamic model expansion with efficient compression.

📊 Dataset Link:

• XRF55 pages: https://aiotgroup.github.io/XRF55/
• XRF55 SDP: http://www.sdp8.org/Dataset?id=705e08e7-637e-49a1-aff1-b2f9644467ae

An overview of the CAREC framework. Each incremental learning session consists of two alternating phases: feature expansion and feature compression. In the expansion phase, a new backbone is trained to incorporate new classes, while in the compression phase, knowledge from multiple backbones is distilled into a compact student model. The process supports lifelong learning without growing model size.

Method	Session 0	Session 1	Session 2	Session 3	Session 4
Baseline	89.67	38.84	26.97	21.83	16.25
iCaRL	90.41	74.09	66.90	63.68	63.49
BiC	86.11	64.91	52.56	50.10	44.35
UCIR	88.07	77.16	69.24	61.73	63.58
BEEF	89.81	73.11	63.46	57.72	53.73
CCS	87.48	75.02	66.59	59.16	60.45
CAREC	89.44	81.36	75.00	63.46	67.84

🏃‍♂️ Running the Code:

1. Navigate to the /exps directory and modify the .json files for different methods (e.g., icarl.json) Key customizable parameters:

 "dataset": "your_dataset_name",  # Dataset configuration
 "device": [0],                 # GPU indices (e.g., [0,1] for multi-GPU)
 "convnet_type": "unet",             # Model architecture
 "memory_size": 1650,             # Size of memory buffer

2. To train the model: Execute the following command to start training:

   python main.py --config ./exp/der.json

3. All training results are automatically saved in the /logs directory

📞 Support

If you encounter any issues or need assistance, feel free to reach out to us.

📖 Related Work

@article{wang2024xrf55,
  title={Xrf55: A radio frequency dataset for human indoor action analysis},
  author={Wang, Fei and Lv, Yizhe and Zhu, Mengdie and Ding, Han and Han, Jinsong},
  journal={Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies},
  volume={8},
  number={1},
  pages={1--34},
  year={2024},
  publisher={ACM New York, NY, USA}
}

@article{fu2024ccs,
  title={CCS: Continuous Learning for Customized Incremental Wireless Sensing Services},
  author={Fu, Qunhang and Wang, Fei and Zhu, Mengdie and Ding, Han and Han, Jinsong and Han, Tony Xiao},
  journal={arXiv preprint arXiv:2412.04821},
  year={2024}
}