README.md

---
library_name: pytorch
license: cc-by-4.0
tags:
- backbone
- android
pipeline_tag: video-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/video_mae/web-assets/model_demo.png)

# Video-MAE: Optimized for Mobile Deployment
## Sports and human action recognition in videos


Video MAE (Masked Auto Encoder) is a network for doing video classification that uses the ViT (Vision Transformer) backbone.

This model is an implementation of Video-MAE found [here](https://github.com/MCG-NJU/VideoMAE).


This repository provides scripts to run Video-MAE on Qualcomm® devices.
More details on model performance across various devices, can be found
[here](https://aihub.qualcomm.com/models/video_mae).


### Model Details

- **Model Type:** Video classification
- **Model Stats:**
  - Model checkpoint: Kinectics-400
  - Input resolution: 224x224
  - Number of parameters: 87.7M
  - Model size: 335 MB

| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|---|---|---|---|---|---|---|---|---|
| Video-MAE | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 208.803 ms | 0 - 40 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 449.531 ms | 9 - 12 MB | FP16 | NPU | [Video-MAE.so](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.so) |
| Video-MAE | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 688.074 ms | 0 - 374 MB | FP16 | NPU | [Video-MAE.onnx](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.onnx) |
| Video-MAE | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 152.411 ms | 0 - 484 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 370.594 ms | 9 - 29 MB | FP16 | NPU | [Video-MAE.so](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.so) |
| Video-MAE | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 520.976 ms | 9 - 294 MB | FP16 | NPU | [Video-MAE.onnx](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.onnx) |
| Video-MAE | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 140.087 ms | 0 - 486 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 363.831 ms | 9 - 424 MB | FP16 | NPU | Use Export Script |
| Video-MAE | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 625.199 ms | 9 - 300 MB | FP16 | NPU | [Video-MAE.onnx](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.onnx) |
| Video-MAE | SA7255P ADP | SA7255P | TFLITE | 2612.002 ms | 0 - 486 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | SA7255P ADP | SA7255P | QNN | 3186.456 ms | 3 - 13 MB | FP16 | NPU | Use Export Script |
| Video-MAE | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 206.797 ms | 0 - 39 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | SA8255 (Proxy) | SA8255P Proxy | QNN | 450.019 ms | 9 - 12 MB | FP16 | NPU | Use Export Script |
| Video-MAE | SA8295P ADP | SA8295P | TFLITE | 351.185 ms | 0 - 495 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | SA8295P ADP | SA8295P | QNN | 668.586 ms | 0 - 17 MB | FP16 | NPU | Use Export Script |
| Video-MAE | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 208.932 ms | 0 - 38 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | SA8650 (Proxy) | SA8650P Proxy | QNN | 449.795 ms | 9 - 12 MB | FP16 | NPU | Use Export Script |
| Video-MAE | SA8775P ADP | SA8775P | TFLITE | 282.557 ms | 0 - 486 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | SA8775P ADP | SA8775P | QNN | 605.641 ms | 9 - 19 MB | FP16 | NPU | Use Export Script |
| Video-MAE | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 2612.002 ms | 0 - 486 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 3186.456 ms | 3 - 13 MB | FP16 | NPU | Use Export Script |
| Video-MAE | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 208.693 ms | 0 - 42 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 449.8 ms | 9 - 12 MB | FP16 | NPU | Use Export Script |
| Video-MAE | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 282.557 ms | 0 - 486 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 605.641 ms | 9 - 19 MB | FP16 | NPU | Use Export Script |
| Video-MAE | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 315.69 ms | 0 - 494 MB | FP16 | NPU | [Video-MAE.tflite](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.tflite) |
| Video-MAE | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 630.074 ms | 9 - 434 MB | FP16 | NPU | Use Export Script |
| Video-MAE | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 472.11 ms | 9 - 9 MB | FP16 | NPU | Use Export Script |
| Video-MAE | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 720.056 ms | 187 - 187 MB | FP16 | NPU | [Video-MAE.onnx](https://huggingface.co/qualcomm/Video-MAE/blob/main/Video-MAE.onnx) |




## Installation


Install the package via pip:
```bash
pip install "qai-hub-models[video-mae]"
```


## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device

Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.

With this API token, you can configure your client to run models on the cloud
hosted devices.
```bash
qai-hub configure --api_token API_TOKEN
```
Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information.



## Demo off target

The package contains a simple end-to-end demo that downloads pre-trained
weights and runs this model on a sample input.

```bash
python -m qai_hub_models.models.video_mae.demo
```

The above demo runs a reference implementation of pre-processing, model
inference, and post processing.

**NOTE**: If you want running in a Jupyter Notebook or Google Colab like
environment, please add the following to your cell (instead of the above).
```
%run -m qai_hub_models.models.video_mae.demo
```


### Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
device. This script does the following:
* Performance check on-device on a cloud-hosted device
* Downloads compiled assets that can be deployed on-device for Android.
* Accuracy check between PyTorch and on-device outputs.

```bash
python -m qai_hub_models.models.video_mae.export
```
```
Profiling Results
------------------------------------------------------------
Video-MAE
Device                          : Samsung Galaxy S23 (13)
Runtime                         : TFLITE                 
Estimated inference time (ms)   : 208.8                  
Estimated peak memory usage (MB): [0, 40]                
Total # Ops                     : 558                    
Compute Unit(s)                 : NPU (558 ops)          
```


## How does this work?

This [export script](https://aihub.qualcomm.com/models/video_mae/qai_hub_models/models/Video-MAE/export.py)
leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
on-device. Lets go through each step below in detail:

Step 1: **Compile model for on-device deployment**

To compile a PyTorch model for on-device deployment, we first trace the model
in memory using the `jit.trace` and then call the `submit_compile_job` API.

```python
import torch

import qai_hub as hub
from qai_hub_models.models.video_mae import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S24")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

```


Step 2: **Performance profiling on cloud-hosted device**

After compiling models from step 1. Models can be profiled model on-device using the
`target_model`. Note that this scripts runs the model on a device automatically
provisioned in the cloud.  Once the job is submitted, you can navigate to a
provided job URL to view a variety of on-device performance metrics.
```python
profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)
        
```

Step 3: **Verify on-device accuracy**

To verify the accuracy of the model on-device, you can run on-device inference
on sample input data on the same cloud hosted device.
```python
input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

```
With the output of the model, you can compute like PSNR, relative errors or
spot check the output with expected output.

**Note**: This on-device profiling and inference requires access to Qualcomm®
AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).




## Deploying compiled model to Android


The models can be deployed using multiple runtimes:
- TensorFlow Lite (`.tflite` export): [This
  tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
  guide to deploy the .tflite model in an Android application.


- QNN (`.so` export ): This [sample
  app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
provides instructions on how to use the `.so` shared library  in an Android application.


## View on Qualcomm® AI Hub
Get more details on Video-MAE's performance across various devices [here](https://aihub.qualcomm.com/models/video_mae).
Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)


## License
* The license for the original implementation of Video-MAE can be found
  [here](https://github.com/MCG-NJU/VideoMAE/blob/main/LICENSE).
* The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)



## References
* [Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://arxiv.org/abs/2203.12602)
* [Source Model Implementation](https://github.com/MCG-NJU/VideoMAE)



## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:[email protected]).