definition.yaml

benchmark:
  DIV2K:
    PSNR: 35.00dB
description: "SESR, super-efficient super resolution\
  \ is a network aims to generate a high-resolution image from a low-resolution input.\n\
  Name was changed by ARM developers when they wrote research paper on their technique.\n\
  The attached int8 fully quantized tflite model achieves 35.00dB PSNR on DIV2K dataset.\
  \ \nThe model takes 1080p input (in YCbCr, i.e., takes a 1x1920x1080x1 tensor as\
  \ input) and outputs 4K images (in YCbCr, i.e., 1x3840x2160x1 output).\n Compatability\
  \ note:\n Please note that SESR is a high-end network operating on 1080p->4K images\
  \ and runtime memory use of this network requires an end system with at least 100MB\
  \ of memory available to ensure successful execution.\n We anticipate the network\
  \ being used in premium devices as part of a camera imaging pipeline providing highest\
  \ quality digital zoom.\n\
  \ Repository for model authors: https://github.com/ARM-software/sesr"
license:
- Apache-2.0
network:
  file_size_bytes: 23680
  filename: SESR_1080p_to_4K_withD2S_full_int8.tflite
  framework: TensorFlow Lite
  hash:
    algorithm: sha1
    value: 5abc5f05202aa1b0b9c34c5a978b6aa0a02f7ec5
  provenance: https://git.research.arm.com/golden-models/super-resolution-model-toolkit/-/tree/shuokai_tf2
  quality: Optimised
network_parameters:
  input_nodes:
  - description: 'Low-resolution input: 1080p (in YCbCr, i.e., take a 1x1920x1080x1
      tensor as input) '
    example_input:
      path: models/superresolution/SESR/tflite_int8/testing_input/net_input
    name: net_input
    shape:
    - 1
    - 1920
    - 1080
    - 1
  output_nodes:
  - description: 'High-resolution input: 4K images (in YCbCr, i.e., 1x3840x2160x1
      output).'
    name: net_output
    shape:
    - 1
    - 3840
    - 2160
    - 1
    test_output_path: models/superresolution/SESR/tflite_int8/testing_output/net_output
operators:
  TensorFlow Lite:
  - ADD
  - BATCH_TO_SPACE_ND
  - CONV_2D
  - RELU
  - TRANSPOSE
paper: https://arxiv.org/abs/2103.09404