# Creating a object detection data set from one folder of several video frames

**URL:** <https://discuss.huggingface.co/t/creating-a-object-detection-data-set-from-one-folder-of-several-video-frames/47168>\
**Category:** Beginners\
**Created:** [July 18, 2023, 8:42am UTC](https://discuss.huggingface.co/t/creating-a-object-detection-data-set-from-one-folder-of-several-video-frames/47168 "2023-07-18T08:42:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![SDK1986](https://avatars.discourse-cdn.com/v4/letter/s/58f4c7/32.png) [@SDK1986](https://discuss.huggingface.co/u/SDK1986)\
**Post date:** [July 18, 2023, 8:42am UTC](https://discuss.huggingface.co/t/creating-a-object-detection-data-set-from-one-folder-of-several-video-frames/47168/1 "2023-07-18T08:42:08Z")

</div>

Hi all,

I have spent the last few weeks trying to get my head around hugging face and its possibilities. I have been unable to figure out how to create a data set from one folder with just under 18000 video frames from several videos. I exported the video frames using CVAT for videos.  
I have one XML file with all the bounding box point data, and I am wanting to create a data set that can then be split to the train test and validate datasets.

Then hopefully figure out how to use the pipeline() function to fine-tune a model. I am stuck and need some guidance on what to do next.

Thank you.

---

<div class="post-metadata">

**Author:** ![SDK1986](https://avatars.discourse-cdn.com/v4/letter/s/58f4c7/32.png) [@SDK1986](https://discuss.huggingface.co/u/SDK1986)\
**Post date:** [August 2, 2023, 2:47am UTC](https://discuss.huggingface.co/t/creating-a-object-detection-data-set-from-one-folder-of-several-video-frames/47168/2 "2023-08-02T02:47:38Z")

</div>

For others who get stuck, I have got to the stage now that I can begin fine-tuning a pre-trained model by following these steps.

1. finding the desired annotation format for the selected pre-trained model. In my case, it was DETR and one JSON file with all the data was required in a specific format that was found by looking at [DETR (huggingface.co)](https://huggingface.co/docs/transformers/model_doc/detr)

2. split the images into train, validate and test datasets. As the hugging face hub only allows 10,000 files per dataset.

3. for each dataset, create a JSON file in the correct format with the frames data. I wrote a bespoke script in python to do this.

4. write a script to fine-tune the selected pre-trained model.