Skip to content

Repository files navigation

Action Detection on MERL Shopping Dataset

MERL Shopping Dataset

The MERL Shopping Dataset contains 106 videos, each of which is a sequence ~2 minutes long. Each video contains several instances of the following 6 actions:

  1. Reach To Shelf - (reach to shelf)
  2. Retract From Shelf - (retract hand from shelf)
  3. Hand In Shelf - (extended period with hand in the shelf)
  4. Inspect Product - (inspect product while holding it in hand)
  5. Inspect Shelf - (look at shelf while not touching and not reaching for the shelf)
  6. None of the above

Steps to create environment

  1. wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
  2. chmod u+x Miniconda3-latest-Linux-x86_64.sh
  3. ./Miniconda3-latest-Linux-x86_64.sh
  4. cd miniconda3/
  5. conda list
  6. source .bashrc
  7. conda create -n tf python=3.7 anaconda
  8. conda activate tf
  9. conda install -c anaconda tensorflow-gpu
  10. conda install -c conda-forge opencv

Steps to run

  1. Download and extract MERL Shopping Dataset - ftp://ftp.merl.com/pub/tmarks/MERL_Shopping_Dataset/
  2. Download Videos_MERL_Shopping_Dataset.zip and unzip it.
  3. Create train, val and test folders inside the unzipped folder for Videos_MERL_Shopping_Dataset.
  4. Move 1st 60 videos (1_1_crop to 20_3_crop) to the train folder.
  5. Move the next 18 videos (21_1_crop to 26_3_crop) to the val folder
  6. Move the remaining 28 videos (27_1_crop to 41_2_crop) to the test folder
  7. Then run the following command on the terminal "./convert_video_to_images.sh Videos_MERL_Shopping_Dataset"
  8. Repeat the above steps 2-6 similarly for Labels_MERL_Shopping_Dataset and then run Generate_output_pickle_files.ipynb.
  9. Activate the conda environment and run ActionRecognition.py on terminal or ActionRecognition.ipynb to run in the jupyter environment
  10. In order to run ResNet Model run Resnet.py

References

  1. Singh, Bharat, et al. "A multi-stream bi-directional recurrent neural network for fine-grained action detection." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016.

Contributors

  1. Rushabh Dharia
  2. Animesh Sagar

About

ENGR E-533 Deep Learning Systems course project

Resources

Stars

7 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages