Datasets:
Added dataset card
Browse files
README.md
CHANGED
@@ -1,3 +1,95 @@
|
|
1 |
-
---
|
2 |
-
license: creativeml-openrail-m
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: creativeml-openrail-m
|
3 |
+
task_categories:
|
4 |
+
- image-segmentation
|
5 |
+
- image-classification
|
6 |
+
- image-feature-extraction
|
7 |
+
language:
|
8 |
+
- en
|
9 |
+
pretty_name: SleetView Agentic Ai Dataset
|
10 |
+
size_categories:
|
11 |
+
- n<1K
|
12 |
+
---
|
13 |
+
# The SleetView Agentic AI Dataset
|
14 |
+
|
15 |
+
The SleetView Agentic AI dataset is a collection of synthetic content automatically generated using Agentic AI
|
16 |
+
|
17 |
+
## Dataset Details
|
18 |
+
|
19 |
+
### Dataset Description
|
20 |
+
|
21 |
+
The images were generated with a collection of models available under the Apache-2.0 or creativeml-openrail-m licenses.
|
22 |
+
To generate this dataset we used our own agentic implementation given the goal of creating a dataset that can be used to research synthetic content detection.
|
23 |
+
As pioneers in the synthetic content detection realm, we think having a varied sampling of synthetic data is important to determine detection efficiency.
|
24 |
+
|
25 |
+
This dataset allows evaluation for the following scenarios:
|
26 |
+
* Varied aspect ratios including multiple resolutions common in digital systems
|
27 |
+
* Landscapes, portraits with one or more characters, pets and other animals, automobiles and architecture
|
28 |
+
* Multiple lighting scenarios including ambient light, spot lights, night time/day time
|
29 |
+
|
30 |
+
Also included in the dataset are segmentation masks and metadata generated with the DETR panoptic model: https://huggingface.co/facebook/detr-resnet-101-panoptic
|
31 |
+
|
32 |
+
- **Shared by:** Mendit.AI
|
33 |
+
- **License:** creativeml-openrail-m
|
34 |
+
|
35 |
+
## Uses
|
36 |
+
|
37 |
+
This dataset can be useful for the following research areas:
|
38 |
+
* Synthetic content detection
|
39 |
+
* Evaluation of the quality of Agentic AI generated synthetic datasets
|
40 |
+
* Image segementation quality based on different composition and lighting scenarios
|
41 |
+
|
42 |
+
### Out-of-Scope Use
|
43 |
+
|
44 |
+
Please refer to the creativeml-openrail-m license for restrictions
|
45 |
+
|
46 |
+
## Dataset Structure
|
47 |
+
|
48 |
+
The dataset contains 248 images with the following structure:
|
49 |
+
* Image
|
50 |
+
* Associated segmentation mask
|
51 |
+
* Segmentation metadata formatted as json
|
52 |
+
|
53 |
+
## Dataset Creation
|
54 |
+
|
55 |
+
### Agentic AI Generation
|
56 |
+
|
57 |
+
An in depth explanation of our approach to agentic generation of synthetic content can be found here: https://menditai.substack.com/p/the-night-the-dataset-appeared-an
|
58 |
+
We opted for a local setup using Ollama and Falcon3 as the LLM powering the agent.
|
59 |
+
Based on our experience with this process we find:
|
60 |
+
* Instruction tuned LLMs that do not including reasoning are the best for this task
|
61 |
+
|
62 |
+
### Annotations [optional]
|
63 |
+
|
64 |
+
Segmentation masks and metadata were automatically generated using the DETR panoptic model: https://huggingface.co/facebook/detr-resnet-101-panoptic
|
65 |
+
|
66 |
+
## Citation [optional]
|
67 |
+
|
68 |
+
**BibTeX:**
|
69 |
+
|
70 |
+
@article{DBLP:journals/corr/abs-2005-12872,
|
71 |
+
author = {Nicolas Carion and
|
72 |
+
Francisco Massa and
|
73 |
+
Gabriel Synnaeve and
|
74 |
+
Nicolas Usunier and
|
75 |
+
Alexander Kirillov and
|
76 |
+
Sergey Zagoruyko},
|
77 |
+
title = {End-to-End Object Detection with Transformers},
|
78 |
+
journal = {CoRR},
|
79 |
+
volume = {abs/2005.12872},
|
80 |
+
year = {2020},
|
81 |
+
url = {https://arxiv.org/abs/2005.12872},
|
82 |
+
archivePrefix = {arXiv},
|
83 |
+
eprint = {2005.12872},
|
84 |
+
timestamp = {Thu, 28 May 2020 17:38:09 +0200},
|
85 |
+
biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib},
|
86 |
+
bibsource = {dblp computer science bibliography, https://dblp.org}
|
87 |
+
}
|
88 |
+
|
89 |
+
@misc{Falcon3,
|
90 |
+
title = {The Falcon 3 Family of Open Models},
|
91 |
+
url = {https://huggingface.co/blog/falcon3},
|
92 |
+
author = {Falcon-LLM Team},
|
93 |
+
month = {December},
|
94 |
+
year = {2024}
|
95 |
+
}
|