Deepen AI vs MakeSense.ai

Deepen AI provides connected data infrastructure for Physical AI, linking sensor calibration, capture, curation, annotation, and validation. Deepen Calibrate aligns camera, LiDAR, radar, and IMU sensors to a shared frame; the site reports under 0.8-pixel reprojection error and sub-1° angular error. Deepen Curate deduplicates sensor data, searches clips and frames, and surfaces rare or long-tail scenarios before labeling. Deepen Annotate handles 2D and 3D labels, including boxes, segmentation, keypoints, point clouds, and multi-frame tracking. Deepen Validate scores scenario coverage, audits and compares labels, prioritizes risky frames for review, and records signed audit trails. REST APIs and SDKs connect products to customer pipelines. Supported formats include ROS 2, MCAP, nuScenes, HDF5, Parquet, COCO, Rosbag, Zarr, KITTI, ORC, Pascal, WebDataset, and ASAM OpenLABEL. Deployment spans managed cloud, VPC, on-premises, and air-gapped environments. Pricing is on request.

MakeSense.ai is a browser-based tool for labeling photos and preparing datasets for computer vision projects. It supports point, line, rectangle, polygon, and label annotations, with model-assisted labeling through YOLOv5, SSD pretrained on COCO, and PoseNet. Users can load their own YOLOv5 models exported to TensorFlow.js format. Export options include CSV, YOLO, VOC XML, VGG JSON, and COCO JSON; imports include YOLO, VOC XML, and COCO JSON. The available format combinations vary by annotation type, and some are not yet supported. The project says images are not sent to a server and AI features run using TensorFlow.js. It works in a browser across operating systems, and documented local and Docker setups are available; Docker serves the application on localhost:3000. The project describes it as suitable for small computer vision deep-learning projects and dataset preparation. It was founded in 2019, is licensed under GPL-3.0, and credits Piotr Skalski as author and copyright holder. The Free plan costs nothing.