Invigilo is Singapore’s leading AI video analytics solution provider for workplace safety. The company uses computer vision models in their video analytics software to enable real-time and continuous monitoring of critical events across different industry verticals such as construction sites, shipyards and manufacturing plants. Examples of this work include detecting people in close proximity to heavy machinery.
“Before using SUPA, we struggled with managing our internal data annotation processes. We’re now seeing better team morale as we are able to move faster and concentrate on model building and deployment.” — Vishnu Saran, CEO at Invigilo
Before SUPA, Invigilo relied on in-house data labelers to create and manage their labeled data. As a result, they were not able to reliably improve their label quality, efficiency and identify edge cases where their models were underperforming. Their team was managing all of their unstructured data through heavily manual workflows, spreadsheets and open-source labeling tools, which had scaling issues with quality and speed. This caused fatigue and took time away from mission-critical model development tasks.
To address these concerns, Invigilo sought a data labeling partner that could deliver quality annotation data with a fast turnaround time and real-time visibility on annotation data quality. SUPA was selected after a thorough evaluation of more than 10 options. The SUPA BOLT platform delivered high-quality annotations within a 24-hour turnaround time, and allowed Invigilo's ML team to review the annotation data as it was being completed by annotators. This real-time feedback mechanism ensured that quality benchmarks were met, and any issues were addressed immediately.
As a result, Invigilo’s team was able to immediately see a 80% gain in annotation delivery speed and a 15% increase in annotation quality. In the future, Invigilo will be looking to enhance its ML-powered video analytics offerings with more capabilities such as automated incident reporting.
.png){kind=avatar}
If there's one thing that makes SUPA stand out, it's their commitment to providing consistent support throughout the data labeling process. The team actively and efficiently engaged with us to ensure any ambiguity in the dataset was cleared up.
Everything from uploading data to seeing it labeled in real time was really cool. This is just way simpler to use compared to Amazon Sagemaker and LabelBox. I was also very impressed with how the platform delivered exactly what we needed in terms of label quality.
I was also able to view the labels as they were being generated, which gave me quick feedback about the label quality, rather than waiting for the whole batch. This replaced my standard manual QA process using external tools like Voxel's Fiftyone, as the labels were clear and easy to parse through in real-time.
The annotators were really quick. I would upload and 5 minutes later - 10 images done. I checked 5 minutes later - 100 images done.
The team at [SUPA] has been very professional & easy to work with since we started our collaboration in 2019. They've provided us with good quality judgments to train, tune, and validate our Search & Recommendations models.
Let us walk you through the entire data labeling experience, from set up to export
Schedule a chat.png)
.png)
.png){kind=small}
.png){kind=small}