SFU-HW-Tracks-v1: Object Tracking Dataset on Raw Video Sequences

TL;DR

The paper introduces SFU-HW-Tracks-v1, a new dataset with annotated object identities for HEVC sequences, enabling evaluation of object tracking performance on uncompressed videos and analysis of compression effects.

Contribution

It provides a comprehensive dataset with ground-truth annotations for object tracking on HEVC sequences, facilitating research on tracking accuracy and video compression impact.

Findings

Dataset includes 13 annotated sequences with object IDs and bounding boxes.

Enables evaluation of object tracking on uncompressed video sequences.

Supports studying the relationship between video compression and tracking performance.

Abstract

We present a dataset that contains object annotations with unique object identities (IDs) for the High Efficiency Video Coding (HEVC) v1 Common Test Conditions (CTC) sequences. Ground-truth annotations for 13 sequences were prepared and released as the dataset called SFU-HW-Tracks-v1. For each video frame, ground truth annotations include object class ID, object ID, and bounding box location and its dimensions. The dataset can be used to evaluate object tracking performance on uncompressed video sequences and study the relationship between video compression and object tracking.