TPDNet

Temporal Prompt Learning with Depth Memory for Video Mirror Detection

Trained model and predicted results

We provide the trained model and predicted results on VMD-D dataset in here.

1. Data Download

The dataset using in this project is VMD-D dataset, which can be downloaded from here.

The downloaded data should be put in the ./VMD folder.

2. Train

The training script is train.py, where we set env to pytorch and num_gpus to 1.

If you want to train the model on multiple GPUs, you can set env to DDP and num_gpus to the number of GPUs you want to use.

3. Predict results

The predict script is validate.py. In this script, we load the trained model (best_model_0.6690.pt) and predict the results on the test set. The predicted results will be saved at ./results/tqdm.

4. Evaluation

After running the predict script, we can evaluate the results by running the eval.py script.

For example:

python eval.py --pred=./results/tqdm