Spaces:
Running
Running
Uploading Trashify V2 box detection model (with data augmentation) app.py
Browse files- README.md +11 -2
- app.py +4 -2
- examples/trashify_example_3.jpeg +0 -0
README.md
CHANGED
|
@@ -18,7 +18,16 @@ Used as example for encouraging people to cleanup their local area.
|
|
| 18 |
|
| 19 |
If `trash`, `hand`, `bin` all detected = +1 point.
|
| 20 |
|
| 21 |
-
|
| 22 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
|
| 24 |
TK - finish the README.md + update with links to materials
|
|
|
|
| 18 |
|
| 19 |
If `trash`, `hand`, `bin` all detected = +1 point.
|
| 20 |
|
| 21 |
+
## Dataset
|
| 22 |
+
|
| 23 |
+
All Trashify models are trained on a custom hand-labelled dataset of people picking up trash and placing it in a bin.
|
| 24 |
+
|
| 25 |
+
The dataset can be found on Hugging Face as [`mrdbourke/trashify_manual_labelled_images`](https://huggingface.co/datasets/mrdbourke/trashify_manual_labelled_images).
|
| 26 |
+
|
| 27 |
+
## Demos
|
| 28 |
+
|
| 29 |
+
* [V1](https://huggingface.co/spaces/mrdbourke/trashify_demo_v1) = Fine-tuned DETR model trained *without* data augmentation.
|
| 30 |
+
* [V2](https://huggingface.co/spaces/mrdbourke/trashify_demo_v2) = Fine-tuned DETR model trained *with* data augmentation.
|
| 31 |
+
* [V3](https://huggingface.co/spaces/mrdbourke/trashify_demo_v3) = Fine-tuned DETR model trained *with* data augmentation (same as V2) with an NMS (Non Maximum Suppression) post-processing step.
|
| 32 |
|
| 33 |
TK - finish the README.md + update with links to materials
|
app.py
CHANGED
|
@@ -131,12 +131,14 @@ demo = gr.Interface(
|
|
| 131 |
],
|
| 132 |
title="🚮 Trashify Object Detection Demo V2",
|
| 133 |
description="""Help clean up your local area! Upload an image and get +1 if there is all of the following items detected: trash, bin, hand.
|
| 134 |
-
|
|
|
|
| 135 |
""",
|
| 136 |
# Examples come in the form of a list of lists, where each inner list contains elements to prefill the `inputs` parameter with
|
| 137 |
examples=[
|
| 138 |
["examples/trashify_example_1.jpeg", 0.25],
|
| 139 |
-
["examples/trashify_example_2.jpeg", 0.25]
|
|
|
|
| 140 |
],
|
| 141 |
cache_examples=True
|
| 142 |
)
|
|
|
|
| 131 |
],
|
| 132 |
title="🚮 Trashify Object Detection Demo V2",
|
| 133 |
description="""Help clean up your local area! Upload an image and get +1 if there is all of the following items detected: trash, bin, hand.
|
| 134 |
+
|
| 135 |
+
The [model](https://huggingface.co/mrdbourke/detr_finetuned_trashify_box_detector_with_data_aug) in V2 has been trained with data augmentation preprocessing (color jitter, horizontal flipping) to improve robustness.
|
| 136 |
""",
|
| 137 |
# Examples come in the form of a list of lists, where each inner list contains elements to prefill the `inputs` parameter with
|
| 138 |
examples=[
|
| 139 |
["examples/trashify_example_1.jpeg", 0.25],
|
| 140 |
+
["examples/trashify_example_2.jpeg", 0.25],
|
| 141 |
+
["examples/trashify_example_3.jpeg", 0.25]
|
| 142 |
],
|
| 143 |
cache_examples=True
|
| 144 |
)
|
examples/trashify_example_3.jpeg
ADDED
|