[Task]: It can be detected when the feature points occupy a certain proportion of the original image？

[xingfenghaizeiwang]

What is the motivation for this task?

I want to detect missing characters on the label

Describe the solution you'd like

dataset:
name: mvtec
format: folder
root: D:\project\anomalib\datasets\MVTec
normal_dir: card/train/good
normal_test_dir: card/test/good
abnormal_dir: card/test/ng
task: segmentation
mask_dir: null
extensions: null
train_batch_size: 32
eval_batch_size: 32
num_workers: 8
image_size:

• 256
• 256
  center_crop: null
  normalization: imagenet
  transform_config:
  train: null
  eval: null
  test_split_mode: none
  test_split_ratio: 0.2
  val_split_mode: synthetic
  val_split_ratio: 0.5
  tiling:
  apply: false
  tile_size: null
  stride: null
  remove_border_count: 0
  use_random_tiling: false
  random_tile_count: 16
  model:
  name: padim
  backbone: wide_resnet50_2
  pre_trained: true
  layers:
• layer1
• layer2
• layer3
  normalization_method: min_max
  input_size:
• 256
• 256
  metrics:
  image:
• F1Score
• AUROC
  pixel:
• F1Score
• AUROC
  threshold:
  method: adaptive
  manual_image: null
  manual_pixel: null
  visualization:
  show_images: false
  save_images: true
  log_images: true
  image_save_path: null
  mode: full
  project:
  seed: 42
  path: result\padim\mvtec\run
  unique_dir: false
  logging:
  logger: []
  log_graph: false
  optimization:
  export_mode: onnx
  trainer:
  enable_checkpointing: true
  default_root_dir: result\padim\mvtec\run
  gradient_clip_val: 0
  gradient_clip_algorithm: norm
  num_nodes: 1
  devices: 1
  enable_progress_bar: true
  overfit_batches: 0.0
  track_grad_norm: -1
  check_val_every_n_epoch: 1
  fast_dev_run: false
  accumulate_grad_batches: 1
  max_epochs: 1
  min_epochs: null
  max_steps: -1
  min_steps: null
  max_time: null
  limit_train_batches: 1.0
  limit_val_batches: 1.0
  limit_test_batches: 1.0
  limit_predict_batches: 1.0
  val_check_interval: 1.0
  log_every_n_steps: 50
  accelerator: auto
  strategy: null
  sync_batchnorm: false
  precision: 32
  enable_model_summary: true
  num_sanity_val_steps: 0
  profiler: null
  benchmark: false
  deterministic: false
  reload_dataloaders_every_n_epochs: 0
  auto_lr_find: false

replace_sampler_ddp: true
detect_anomaly: false
auto_scale_batch_size: false
plugins: null
move_metrics_to_cpu: true
multiple_trainloader_mode: max_size_cycle

Additional context

No response

[blaz-r]

Hello. Could you clarify this issue please? The title and request don't match so I'm not sure what is the problem.

[samet-akcay]

I agree with @blaz-r. @xingfenghaizeiwang, could you please clarify the issue you are having?

[xingfenghaizeiwang]

你好。你能澄清一下这个问题吗？标题和请求不匹配，所以我不确定问题出在哪里。

Unable to identify the characters on the card indeed

[xingfenghaizeiwang]

我同意.，您能否澄清您遇到的问题？

Unable to identify the characters on the card indeed.I'm not sure how small a target can be identified

[samet-akcay]

This sounds more like a performance problem for a specific use-case. @blaz-r is working on a cool feature that you could try once it is merged. We will provide an update here