Phoenix - Your AI AssistantWe require dataset labelling with high accuracy
- Posted:
- Proposals: 28
- Remote
- #4289844
- OPPORTUNITY
- Awarded
Frankfurt am Main 
Description
Experience Level: Expert
Hi,
We need a dataset of images labelled with the following labels with a very high accuracy rate. We will need to import the labels into our Google Vertex AI model.
Currently the dataset consists of around 500 images and polygons will need to be drawn around the following objects in order for us to train the model.
- Bodies of water (pools of water only), wet surfaces only should not be labelled
- Birds
- Groups of leaves (not just one leaf)
We need a dataset of images labelled with the following labels with a very high accuracy rate. We will need to import the labels into our Google Vertex AI model.
Currently the dataset consists of around 500 images and polygons will need to be drawn around the following objects in order for us to train the model.
- Bodies of water (pools of water only), wet surfaces only should not be labelled
- Birds
- Groups of leaves (not just one leaf)
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
Could you clarify if you have a preferred tool or format for labeling the images to ensure compatibility with Google Vertex AI? Additionally, are there any specific accuracy guidelines or validation steps you’d like followed for the labeled dataset?
-
write down all possible question?
For importing into Google Vertex AI, are there specific file formats, coordinate systems, or naming conventions we should follow?
Would you prefer tight, pixel-perfect polygons, or would more simplified, bounding polygons around the objects be acceptable?
Are you looking for all bird species to be labeled, or only specific types?