mean Average Precision - This code evaluates the performance of your neural net for object recognition.

Annotate quickly images.

Python-based optical flow toolkit for existing popular dataset

Create 3D labelled bounding boxes in RViz

Easy description to run and evaluate Lego-LOAM with KITTI-data

Face Video Segmentation - Face segmentation ground truth from videos

This project allows the alignment and correction of LiDAR-based SLAM session data with a reference map or another session, also the retrieval of 6-DoF poses with accuracy of up to 3 cm given an accurate TLS point cloud as a reference map (this map should be accurate at least regarding the position of permanent elements such as walls and columns).

OCR & Ground Truth Resources

SimViz contains tools and resources for authoring and executing autonomous vehicle simulations on roadways and city scapes by using map import, scene creation, formatting of ground truth data, and creating spline based roads.

Easy description to run and evaluate A-LOAM with KITTI-data

Ergonomic line-by-line transcription of scanned text.

Annotated Beethoven Corpus (ABC): A dataset of harmonic analyses with standardized labels

Ground Truth Resources for the HTR of patrimonial documents

A framework for benchmarking clustering algorithms

Resolves severe noise in the widely spread CASIA2.0 dataset ground-truth for Image Manipulation Detection

Annotate quickly images.

Convenient image annotation tool API

SOFT-CSRNET : Counting people in drone video footage

A novel approach to evaluate community detection algorithms on ground truth

Transcription corpora for training HTR models for medieval manuscripts from the 12th to the 15th century.