--- library_name: pytorch license: other tags: - android pipeline_tag: image-to-image --- ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/sesr_m5/web-assets/model_demo.png) # SESR-M5: Optimized for Mobile Deployment ## Upscale images in real time SESR M5 performs efficient on-device upscaling of images. This model is an implementation of SESR-M5 found [here](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/sesr). This repository provides scripts to run SESR-M5 on Qualcomm® devices. More details on model performance across various devices, can be found [here](https://aihub.qualcomm.com/models/sesr_m5). ### Model Details - **Model Type:** Model_use_case.super_resolution - **Model Stats:** - Model checkpoint: sesr_m5_3x_checkpoint - Input resolution: 128x128 - Number of parameters: 343K - Model size (float): 1.32 MB - Model size (w8a8): 395 KB | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model |---|---|---|---|---|---|---|---|---| | SESR-M5 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 10.517 ms | 3 - 121 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 10.533 ms | 0 - 116 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 2.811 ms | 0 - 140 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 2.789 ms | 0 - 138 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 1.877 ms | 0 - 3 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 1.86 ms | 0 - 3 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 2.3 ms | 0 - 3 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx.zip) | | SESR-M5 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 3.065 ms | 0 - 118 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 3.034 ms | 0 - 117 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 10.517 ms | 3 - 121 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 10.533 ms | 0 - 116 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 3.474 ms | 0 - 124 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 3.465 ms | 0 - 123 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 3.065 ms | 0 - 118 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 3.034 ms | 0 - 117 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.274 ms | 0 - 133 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 1.274 ms | 0 - 138 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 1.593 ms | 0 - 111 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx.zip) | | SESR-M5 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 1.026 ms | 0 - 123 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 1.026 ms | 0 - 123 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 1.259 ms | 0 - 93 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx.zip) | | SESR-M5 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 0.881 ms | 0 - 122 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.tflite) | | SESR-M5 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 0.881 ms | 0 - 120 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 1.083 ms | 0 - 92 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx.zip) | | SESR-M5 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 2.096 ms | 0 - 0 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.dlc) | | SESR-M5 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.253 ms | 8 - 8 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5.onnx.zip) | | SESR-M5 | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | TFLITE | 6.858 ms | 1 - 122 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | QNN_DLC | 6.799 ms | 0 - 122 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Dragonwing Q-6690 MTP | Qualcomm® QCM6690 | ONNX | 30.347 ms | 38 - 52 MB | CPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | TFLITE | 2.466 ms | 0 - 3 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | QNN_DLC | 2.749 ms | 2 - 4 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Dragonwing RB3 Gen 2 Vision Kit | Qualcomm® QCS6490 | ONNX | 42.247 ms | 38 - 41 MB | CPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 1.945 ms | 1 - 117 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.906 ms | 0 - 116 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.947 ms | 0 - 131 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.941 ms | 0 - 132 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.617 ms | 0 - 3 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.602 ms | 0 - 2 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 0.968 ms | 0 - 3 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 3.377 ms | 0 - 116 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 0.825 ms | 0 - 116 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 1.945 ms | 1 - 117 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.906 ms | 0 - 116 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.407 ms | 0 - 122 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.403 ms | 0 - 122 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 3.377 ms | 0 - 116 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 0.825 ms | 0 - 116 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.466 ms | 0 - 135 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.466 ms | 0 - 134 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.666 ms | 0 - 110 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.379 ms | 0 - 121 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.379 ms | 0 - 121 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.521 ms | 0 - 97 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 0.82 ms | 0 - 121 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.811 ms | 0 - 121 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 26.983 ms | 38 - 54 MB | CPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 0.323 ms | 0 - 119 MB | NPU | [SESR-M5.tflite](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.tflite) | | SESR-M5 | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 0.318 ms | 0 - 119 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 0.473 ms | 0 - 95 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | | SESR-M5 | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.725 ms | 0 - 0 MB | NPU | [SESR-M5.dlc](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.dlc) | | SESR-M5 | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.919 ms | 3 - 3 MB | NPU | [SESR-M5.onnx.zip](https://huggingface.co/qualcomm/SESR-M5/blob/main/SESR-M5_w8a8.onnx.zip) | ## Installation Install the package via pip: ```bash pip install qai-hub-models ``` ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.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://workbench.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.sesr_m5.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.sesr_m5.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.sesr_m5.export ``` ## How does this work? This [export script](https://aihub.qualcomm.com/models/sesr_m5/qai_hub_models/models/SESR-M5/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.sesr_m5 import Model # Load the model torch_model = Model.from_pretrained() # Device device = hub.Device("Samsung Galaxy S25") # 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 Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup). ## Run demo on a cloud-hosted device You can also run the demo on-device. ```bash python -m qai_hub_models.models.sesr_m5.demo --eval-mode on-device ``` **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.sesr_m5.demo -- --eval-mode on-device ``` ## 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 SESR-M5's performance across various devices [here](https://aihub.qualcomm.com/models/sesr_m5). Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/) ## License * The license for the original implementation of SESR-M5 can be found [here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.pdf). ## References * [Collapsible Linear Blocks for Super-Efficient Super Resolution](https://arxiv.org/abs/2103.09404) * [Source Model Implementation](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/sesr) ## 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:ai-hub-support@qti.qualcomm.com).