robust

A Robust and Efficient Framework for Sports-Field Registration

This is an re-implementation of A Robust and Efficient Framework for Sports-Field Registration.

Pretrained Models

1. Download pretrained weight on WorldCup dataset.
2. Now the pretrained models would place in checkpoints.
3. Download public WorldCup dataset.
4. Now the WorldCup dataset would place in dataset/soccer_worldcup_2014.

Evaluation

Evaluation command

python test.py <path/param_text_file>

param_text_file as follows,

• exp_robust.txt: download pretrained weight first and place in checkpoints. Set --train_stage to 0 for testing on WorldCup test set or set --train_stage to 1 on TS-WorldCup test set and set --sfp_finetuned to False.
• exp_robust_finetuned.txt: download pretrained weight first and place in checkpoints. Set --train_stage to 1 and --sfp_finetuned to True for testing finetuned results on TS-WorldCup test set.

We will save heatmap results and corresponding homography matrix into /checkpoints/path of experimental name, which set --name in param_text_file.

Note:

• robust_worldcup_testset_dilated: the preprocess results for predicting on WorldCup dataset and would place in dataset/soccer_worldcup_2014/soccer_data.
• SingleFramePredict_with_normalized: the preprocess results for predicting on TS-WorldCup dataset and would place in dataset/WorldCup_2014_2018.
• SingleFramePredict_finetuned_with_normalized: the preprocess results for finetuning on TS-WorldCup dataset and would place in dataset/WorldCup_2014_2018.

Train model

Train command

python train.py <path/param_text_file>

param_text_file as follows,

• opt_robust.txt: download pretrained weight first and place in checkpoints. Set --train_stage to 0 and --trainset to train_val for training on WorldCup train set.
• opt_robust_finetuned.txt: download pretrained weight first and place in checkpoints. Set --train_stage to 1 and --trainset to train for finetuning on WorldCup train set.

We will save visualize results and weights into /checkpoints/path of experimental name, which set --name in param_text_file.

Note: Please check the following arguments to set correct before training every time.

• --gpu_ids
• --name
• --train_stage and --trainset
• --ckpt_path
• --train_epochs and --step_size

Details refer to options.py.