Home

Awesome

PREGO: online mistake detection in PRocedural EGOcentric videos (CVPR 2024)

| PREGO paper [CVPR 2024] | TI-PREGO paper [arXiv]

Index

  1. Introduction
  2. News
  3. Preparation
  4. Usage
  5. Reference

Introduction

This repo hosts the official PyTorch implementations of the IEEE/CVF Computer Vision and Pattern Recognition (CVPR) '24 paper PREGO: online mistake detection in PRocedural EGOcentric videos and of the follow-up paper TI-PREGO: Chain of Thought and In-Context Learning for Online Mistake Detection in PRocedural EGOcentric Videos.

PREGO is the first online one-class classification model for mistake detection in procedural egocentric videos. It uses an online action recognition component to model current actions and a symbolic reasoning module to predict next actions, detecting mistakes by comparing the recognized current action with the expected future one. We evaluate this on two adapted datasets, Assembly101-O and Epic-tent-O, for online benchmarking of procedural mistake detection.

teaser_image

News

[2024-12-01] Uploaded the recognition branch.

[2024-11-12] Uploaded the script for the prediction aggregation strategy described in [TI-PREGO].

[2024-11-12] Uploaded the TSN features for Assembly101-O and Epic-tent-O [GDrive].

[2024-11-04] Published the follow-up paper [TI-PREGO].

[2024-06-20] Presented PREGO at #CVPR2024.

[2024-06-16] Uploaded the anticipation branch.

Preparation

Data

The TSN features of the Assembly101-O and Epic-tent-O datasets can be downloaded here: [GDrive]. The folder follows the structure described in MiniROAD:

PREGO
|
|__________ Assembly101-O
|           |
|           |__________ rgb_anet_resnet50
|           |           |
|           |           |_________nusar-2021_action_both_9011-b06b_9011_user_id_2021-02-01_154253.npy
|           |           |_________...
|           |__________ rgb_as_flow
|           |           |
|           |           |_________nusar-2021_action_both_9011-b06b_9011_user_id_2021-02-01_154253.npy
|           |           |_________...
|           |__________ target_perframe
|                       |
|                       |_________nusar-2021_action_both_9011-b06b_9011_user_id_2021-02-01_154253.npy
|                       |_________...
|__________ Epic-tent-O
            |
            |__________ rgb_anet_resnet50
            |           |
            |           |_________annotations_1.npy
            |           |_________...
            |__________ rgb_as_flow
            |           |
            |           |_________annotations_1.npy
            |           |_________...
            |__________ target_perframe
                        |
                        |_________annotations_1.npy
                        |_________...

LLAMA

To run our anticipation step with LLAMA, you must be granted access to the models by Meta here. Place them wherever you like, and recall to update the paths whenever necessary, as in step_anticipation/scripts/anticipation.sh.

Environment

You can choose between creating a conda or virtualenv environment, as you prefer

# conda
conda create -n prego python=3.10
conda activate prego

# virtualenv
python3.10 -m venv .venv
source .venv/bin/activate

Then, install the requirements

pip install -r requirements.txt

Install unsloth following the instructions here.

Usage

Step Recognition

For more detaila regarding the Step Recognition branch, you can refer to the official implementation of MiniROAD here.

To run the training on Assembly101-O for example, use the command

python step_recognition/main.py --config step_recognition/configs/miniroad_assembly101-O.yaml

that will save the checkpoints in the folder step_recognition/checkpoint/miniROAD/Assembly101-O.

At this point, you can use the checkpoint for evaluation and it will save predictions frame by frame as a JSON file in the folder output_miniROAD using the command

python step_recognition/main.py --config step_recognition/configs/miniroad_assembly101-O.yaml --eval <checkpoint_path>

Data Aggregation

The utils/aggregate.py script handles the data aggregation process. This script is responsible for aggregating predictions and ground truth data and saving the results to a JSON file.

To run the data aggregation script, use the following command using as input the JSON that was created in the section Step Recognition:

python utils/aggregate.py <input_path> <output_path>

Example

python utils/aggregate.py data/input.json data/output/aggregated_data.json

Step Anticipation

Data Preparation

Description of the steps needed to prepare the data for the Step Anticipation branch.

Step Recognition predictions:

{
    "nusar-2021_action_both_9044-a08_9044_user_id_2021-02-05_154403": {
        "pred": [
            39,
            37,
            74,
            39,
            37
        ],
        "gt": [
            37,
            80,
            39,
            29,
            85
        ]
    },
...
}

Context prompt:

Parameters

Description of the parameters that can be added to the step_anticipation/scripts/anticipation.sh script.

Run

cd step_anticipation
./scripts/anticipation.sh

Reference

If you find our code or paper to be helpful, please consider citing:

@InProceedings{Flaborea_2024_CVPR,
    author    = {Flaborea, Alessandro and di Melendugno, Guido Maria D'Amely and Plini, Leonardo and Scofano, Luca and De Matteis, Edoardo and Furnari, Antonino and Farinella, Giovanni Maria and Galasso, Fabio},
    title     = {PREGO: Online Mistake Detection in PRocedural EGOcentric Videos},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2024},
    pages     = {18483-18492}
}
@misc{plini2024tipregochainthoughtincontext,
      title={TI-PREGO: Chain of Thought and In-Context Learning for Online Mistake Detection in PRocedural EGOcentric Videos}, 
      author={Leonardo Plini and Luca Scofano and Edoardo De Matteis and Guido Maria D'Amely di Melendugno and Alessandro Flaborea and Andrea Sanchietti and Giovanni Maria Farinella and Fabio Galasso and Antonino Furnari},
      year={2024},
      eprint={2411.02570},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2411.02570}, 
}