Dataset Viewer

Split (1)

train

The dataset viewer is not available for this split.

Cannot load the dataset split (in streaming mode) to extract the first rows.

Error code:   StreamingRowsError
Exception:    UnidentifiedImageError
Message:      cannot identify image file <_io.BytesIO object at 0x7f49c49da520>
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                         ^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/utils.py", line 77, in get_rows
                  rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2567, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2103, in __iter__
                  batch = formatter.format_batch(pa_table)
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/formatting/formatting.py", line 472, in format_batch
                  batch = self.python_features_decoder.decode_batch(batch)
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/formatting/formatting.py", line 234, in decode_batch
                  return self.features.decode_batch(batch, token_per_repo_id=self.token_per_repo_id) if self.features else batch
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/features/features.py", line 2254, in decode_batch
                  decode_nested_example(self[column_name], value, token_per_repo_id=token_per_repo_id)
                File "/usr/local/lib/python3.12/site-packages/datasets/features/features.py", line 1508, in decode_nested_example
                  return schema.decode_example(obj, token_per_repo_id=token_per_repo_id) if obj is not None else None
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/features/image.py", line 190, in decode_example
                  image = PIL.Image.open(bytes_)
                          ^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/PIL/Image.py", line 3498, in open
                  raise UnidentifiedImageError(msg)
              PIL.UnidentifiedImageError: cannot identify image file <_io.BytesIO object at 0x7f49c49da520>

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Autofollow Driving Dataset

English | 中文

Dataset Description

This dataset is part of the vision-follow-car project — an ideal hands-on project for learning AI: from data collection and model training to real-time deployment on embedded hardware (Jetson Orin Nano + STM32).

Dataset Summary

Autofollow is an end-to-end imitation learning dataset for autonomous following control. The vehicle learns a "follow when target present, stop when absent" policy based on whether a valid follow target exists in the camera view.

Task: End-to-end imitation learning
Input: 5 consecutive frames + 5 speed values
Output: steering, throttle, brake, target_valid (4-D control)
Frame rate: ~10 Hz, 5 frames ≈ 0.5 s

Supported Tasks

Regression: steer, throttle
Binary classification: brake, target_valid (whether a valid follow target is present)

Languages

English / 中文

Data Structure

Data Instances

Each sample is a 5-frame temporal window with corresponding control labels.

Data Splits

Split	Path	Description
main1	`main1/`	Primary driving data, target-present frames, may include short target-absent tail
main2	`main2/`	Same as main1
no_target	`no_target/`	Target-absent data, labels: steer=0, throttle=0, brake=1, target_valid=0

Data Fields

controls.csv (main1 / main2):

Field	Type	Description
frame_idx	int	Frame index, must match image filename (e.g. 000302.jpg → 302)
image_path	str	Relative path, e.g. `frames/000123.jpg`
steer	float	Steering [-1, 1]
throttle	float	Throttle [0, 1]
brake	float	Brake [0, 1]
speed	float	Current speed
target_valid	int	1=valid follow target present, 0=absent or should not follow (required)
gear, ts, seq, ts_ms, raw	-	Optional, for debugging

Directory structure:

main1/
├── frames/           # Images: 000000.jpg, 000001.jpg, ...
└── controls.csv
main2/
├── frames/
└── controls.csv
no_target/
├── frames/
└── controls.csv

Sample Construction Rules

main1 / main2: Build samples only for 5-frame windows with consecutive frame_idx (difference 1). Labels are determined by the last frame's target_valid.
no_target: Fixed labels: steer=0, throttle=0, brake=1, target_valid=0.

Dataset Creation

Source Data

Collected from real-world or simulation driving, including front-facing camera and speed measurements.

Preprocessing

Images named by frame index (e.g. 000000.jpg)
controls.csv must contain the target_valid column
Use companion tool tool/verify_dataset.py to verify photo–CSV correspondence

Limitations

Dataset size and generalization depend on the collection environment
Domain adaptation or extra validation is recommended before deployment

Citation

If you use this dataset, please cite the source. Project & training code: vision-follow-car.

License

[Specify your license, e.g. MIT / CC BY / Custom]

Downloads last month: 18

Total file size:

2.97 GB