Classify, Detect and Identify
LTU uses deep learning models to classify large batches of images, detect objects in an image, and identify the object detected in an image. Our computer vision processes aim to provide our clients with bespoke, fast and accurate image recognition.
LTU recognises retail products, decorated objects, pictures, book covers, textbooks, art paintings, and logos, and more
Deep Learning & Transfer Learning
Image Matching combined with Transfer Learning
The DNA signature developed by LTU is also robust to the multiple alterations of requests sent from a mobile device due to changes in scale, shooting angles, motion blur, compressions and sometimes low quality optics.
Research and development is a significant part of LTU’s investments. Based on the Transfer Learning principles, LTU implements image classification tailored for specific needs and optimizes the results returned by soliciting several image analysis algorithms. Once results are returned, this automated optimization loop tends to get as close as possible to the ground truth when the repository is poor quality or the queries are altered by exogenous factors.
Detection & Identification
Create your own dataset
Machine learning algorithms are typically designed to address specific tasks. Transfer learning is a popular method in computer vision as it allows us to build accurate models in great time efficiency by leveraging the use of pre-trained models. A pre-trained model is a model that was trained on a large benchmark dataset to solve a problem similar to the one that we want to solve.
LTU interface allows you to drag and drop your dataset and annotate each image using your specific terminology.