Object detection and tracking are important in many computer vision applications including activity recognition, automotive safety, and surveillance. Detection of an object entails not only finding the class of object, but also localizing the extent of an object in an image. Keypoint detection, numbered landmarks, polygons, instance and pixel-level annotations are some of the more popular annotation types applied to object detection algorithms that CVEDIA supports.
Our platform allows you to ingest your images in a standardized manner allowing you to prepare and manage your image dataset efficiently. But we go beyond that by providing advanced tools to improve your workflow. Search, browse and filter every image with our online productivity tools, allowing you to customize every parameter for input.
Flexible augmentations are done in real-time and online. Our Metadata Query Language (MQL) editor allows for you to filter images based on metadata. Create custom collections from metadata filters. Export only the selected annotated images. No need to download GB’s or TB’s of data and save on storage space.
We provide you with powerful annotation tools to train your model to detect a variety of phenomena including the detection of animals for crop protection; person detection for trespassing; detection of illegally parked vehicles; track crop quality and infestation to anticipate crop shortages Real-time infection source can be revealed with high resolution images provided by a drone. Crop production can also be used to inform hedge funds of commodity prices to eliminate margin of error. Our platform also provides annotation tools for image retrieval applications, as well as API B2B room categorization for room/amenities detection.
Our labeling system is highly configurable and multi-layer aware. This means whether you are cropping, labeling or producing segmentation maps you will have the ability to do this in multiple layers and select which one(s) to export in the end. Related labels that are grouped together can be assigned as a whole to a single annotation layer. For example, you can produce a semantic segmentation map of a street view in one layer and have objects masses and direction vectors in the 2nd and 3rd layer.
Labels also serve as templates where you can pre-define additional attributes to be bestowed on an annotated image or polygon. During export, these values can be included as individual fields or combined to form multi-layer segmentation maps.
The easiest method for creating segmentation map annotations is through polygon partitioning. By drawing intersecting splines on top of your source imagery you are effectively cutting the predefined area into many sub-polygons (aka segments). This is a highly efficient method of generating Pixel-dense segmentation maps, as the class for every single pixel is defined from the start, meaning you will not have to manually align the polygons for 2 segments running parallel to each other.
Alternatively, if you are not looking for Pixel-dense representations, but instead want to annotate only certain types of objects, you will have the ability to draw closed polygons. Since these often encompass a whole object including its occluded parts (e.g. car behind a lamp post) we have added the ability to control the Z-index (depth) of the objects. The Z-index defines which segment is closer to the camera and should be rendered last when outputting segmentation map images.
When dealing with irregular shaped or tiny objects, polygons are not always the best solution. For these cases we offer various painting tools that let you easily mark areas based on color or edge detections. Like our landmark tools these resulting annotations can be drawn on top or onto a blank canvas with fully configurable intensity values.
Using our simple point-and-click interface you can assign landmarks to either source imagery or to crops created out of bounding boxes. The value, size and smoothness of the landmark is not set at annotation time, instead it is freely configurable as an augmentation during export allowing for maximum flexibility when experimenting.
Additionally, the landmarks can be drawn onto a blank canvas serving as a ground truth or rendered directly on top of the source images.
An alternative way of using our landmark annotation tool is by predefining their count and order in which to be assigned. This enables the creation of facial landmarks and other annotations that require a fixed feature output size.
In many ways bounding boxes have similar traits to polygons found in segmentation maps. The main difference is the fact they are constrained to either rectangular or square shape. On export they become interchangeable and polygons can automatically be enclosed by a bounding box fitting the polygon's maximum extents. The residual area outside the polygon can be filled through various methods: gaussian noise, background, solid color, opacity etc.
2D bounding boxes are created by first setting the aspect ratio constraints (e.g. 1:1 for square) and dragging the box on top of your image. Multiple labels and attributes can be automatically assigned to a single bounding box by using the label templates or multi-selecting labels. The entire process can be performed in multiple freely definable layers.
Bounding boxes are treated by our system as unique new images and can therefore undergo the same annotation as the original source image. For example, a cropped bounding box on a satellite image or whole slide image can be used as a starting for a segmentation map.