seq_id
stringlengths 5
18
| bbox
listlengths 1.61k
7.22k
|
|---|---|
airliner_1
| [{"t":0.042,"lx":627.9099731445312,"ly":254.38999938964844,"w":59.790000915527344,"h":58.36999893188(...TRUNCATED)
|
arrow_1
| [{"t":0.042,"lx":1011.0,"ly":107.0,"w":205.0,"h":138.0},{"t":0.044,"lx":1007.0,"ly":107.099998474121(...TRUNCATED)
|
badminton_juggle_1
| [{"t":0.006,"lx":477.29998779296875,"ly":361.8999938964844,"w":31.799999237060547,"h":42.29999923706(...TRUNCATED)
|
badminton_juggle_2
| [{"t":0.006,"lx":324.7300109863281,"ly":222.6999969482422,"w":47.56999969482422,"h":37.7999992370605(...TRUNCATED)
|
badminton_racket_1
| [{"t":0.006,"lx":356.9800109863281,"ly":187.6999969482422,"w":31.43000030517578,"h":140.869995117187(...TRUNCATED)
|
badminton_racket_3
| [{"t":0.006,"lx":353.8999938964844,"ly":449.30999755859375,"w":87.0999984741211,"h":102.699996948242(...TRUNCATED)
|
badminton_racket_4
| [{"t":0.006,"lx":689.0,"ly":417.0,"w":49.0,"h":169.0},{"t":0.008,"lx":691.0,"ly":424.6000061035156,"(...TRUNCATED)
|
badminton_racket_5
| [{"t":0.006,"lx":561.0,"ly":371.0,"w":137.0,"h":59.0},{"t":0.008,"lx":570.0999755859375,"ly":368.0,"(...TRUNCATED)
|
badminton_racket_6
| [{"t":0.006,"lx":547.9299926757812,"ly":461.07000732421875,"w":41.0,"h":140.0},{"t":0.008,"lx":553.1(...TRUNCATED)
|
bird_2
| [{"t":0.042,"lx":777.9199829101562,"ly":295.7799987792969,"w":31.18000030517578,"h":55.7099990844726(...TRUNCATED)
|
ESOT500: A High-Frequency Dataset for Event-Driven Perception
Introduction
ESOT500 is a high-frequency annotated event-based single object tracking dataset. It was created to demonstrate the STARE (STream-based lAtency-awaRe Evaluation) framework, enabling rigorous assessment of event-driven perception models' real-time capabilities.
This dataset is introduced in the paper: Bridging the Latency Gap with a Continuous Stream Evaluation Framework in Event-Driven Perception.
Key Features
- 500 Hz Annotations: Provides temporal dense, time-aligned ground truth bounding boxes at 500 Hz, accurately capturing high-dynamic object motion and mitigating temporal aliasing.
- Dual Resolutions: Includes two subsets to test model robustness at different scales:
ESOT500-L: Low-resolution (346x260)ESOT500-H: High-resolution (1280x720)
- Diverse Scenarios: Covers a wide range of indoor/outdoor scenes, object classes, and challenging conditions like high speed, motion blur, and occlusion.
- Continuous-Stream Ready: Designed to evaluate models on continuous event streams, moving beyond the conventional frame-based paradigm.
Dataset Structure
The ESOT500 dataset is organized into two main configurations, ESOT500-L and ESOT500-H. Each configuration contains the following directories:
aedat4/: Contains the raw event stream data in.aedat4format for each sequence.anno_t/: Contains the corresponding 500 Hz time-aligned annotations in.txtformat. Each line in the annotation file represents[timestamp, x, y, width, height].- Split Files:
train.txt: The primary training split.test.txt: The primary testing split.train_additional.txt/test_additional.txt: Additional splits for extended evaluation.*test_challenging.txt: A subset of challenging sequences from the test set.cas.txt: A subset of sequences particularly suited for evaluating Context-Aware Sampling strategies.
Note: The additional split of ESOT500-L was recorded with a slight de-focus, leading to some target blur which could affect tracking performance. Consequently, it was discarded from the primary experimental settings in our paper.
And you can directly download the compressed files at ESOT500/warped.
- Downloads last month
- 179