Datasets:

gemixin
/

touch-ex

---
language:
- en
license: cc-by-4.0

pretty_name: Touch-Ex

task_categories:
- image-classification
- feature-extraction

tags:
- tactile-sensing
- visuo-tactile
- robotics
- digit-sensor
- multimodal
- tactile-language
- tactile-perception
- representation-learning
- force-aware-perception
---
# Touch-Ex: A Region-Level, Force-Annotated Visuo-Tactile Dataset

Developed by Gemma McLean and supervised by Dr Daniel Hao, University of Leicester.

## Dataset Summary

Touch-Ex (Touch Exploration) is a region-level, force-annotated visuo-tactile dataset collected using a DIGIT vision-based tactile sensor and a collection of common UK household objects. The dataset was designed to support research in tactile representation learning, force-aware perception, object-region recognition, temporal tactile modelling, tactile-language alignment and in particular, active tactile exploration.

The dataset contains 140,400 visuo-tactile frames collected across 1,170 tactile interactions from 26 physical objects spanning 18 object categories. The objects cover varied materials, textures, geometries, and compliance levels.

Touch-Ex is divided into four predefined splits:

| Split | Physical objects | Purpose |
| --- | ---: | --- |
| `train` | 14 | Model training, validation, and in-distribution testing using interaction-level partitions |
| `validation_unseen_matched` | 4 | Model and configuration selection using new physical instances of known categories |
| `test_unseen_matched` | 4 | Final evaluation on new physical instances of known categories |
| `test_unseen_related` | 4 | Final analysis of transfer to unfamiliar, tactually related object categories |

Each dataset sample corresponds to a single tactile frame and includes image data, force measurements, interaction metadata, semantic descriptions, and structured tactile attributes.

The training split was designed so that each object contributes an equal number of tactile interactions regardless of the number of annotated regions, while other splits are intentionally smaller.

## Example Samples

| image | object | region | force_level | motion | material | hardness | description |
|--------------|---------|---------|---------|---------|---------|---------|---------|
| ![](images/examples/hammer_handle_sliding_fl3.jpg) | hammer | handle | 3 | sliding | rubber | medium-hard | An elongated, solid rubber-coated gripping area featuring small, evenly spaced circular dimples that provide a repeating texture. |
| ![](images/examples/scissors_blades_rotation_fl2.jpg) | scissors | blades | 2 | rotation | metal | hard | A pair of elongated, flat metallic blades with polished surfaces that gradually taper to sharp outer edges and are joined at a raised pivot screw. |
| ![](images/examples/tv_remote_buttons_sliding_fl1.jpg) | tv_remote | buttons | 1 | sliding | rubber | soft-medium | Small, compact raised rubber features, including rectangles, squares, and circles, that provide springy resistance when pressed. |

*Touch-Ex samples showing tactile images and their key annotations.*

---

## Dataset Modality

Following standard terminology for optical tactile sensors, "visuo-tactile" refers strictly to the internal image data captured by the DIGIT sensor's embedded camera. This dataset intentionally excludes external RGB camera views of the objects to prioritise compactness and pure tactile-language representation learning.

---

## Why Touch-Ex?

Touch-Ex was designed to address several limitations of existing visuo-tactile datasets. While many tactile datasets focus primarily on object identity or contact localisation, Touch-Ex introduces region-level annotations, force-level annotations, and semantic tactile descriptions within a unified benchmark.

Touch-Ex combines:
- Region-level object annotations
- Force-level tactile annotations
- Sequential tactile interactions
- Controlled exploratory motions
- Natural-language tactile descriptions
- Structured semantic attributes

In addition to standard in-distribution evaluation, Touch-Ex separates generalisation validation from final testing. Matched-instance validation supports model and configuration selection, while separate matched-instance and related-category test splits support final assessment of transfer beyond the training objects. This enables systematic evaluation of object-level generalisation and the extent to which learned representations capture transferable tactile properties rather than object identity alone.

---

## Data Collection

Touch-Ex was collected using a DIGIT vision-based tactile sensor operated manually by a human data collector, moving and orienting the DIGIT and the object as needed. 

To enable force-aware data collection, a Grove Round Force Sensor (FSR402) was attached to the rear of the DIGIT sensor and connected to a Raspberry Pi Pico via a Grove Shield. 

During data collection, the operator applied pressure to the rear of the sensor using their index finger (stabilising either side with their thumb and middle finger) while bringing the DIGIT into contact with the target object. The rear-mounted FSR measured this applied force, providing an estimate of how strongly the sensor was pressed against the object surface.

Data acquisition was performed using a custom collection tool developed specifically for Touch-Ex: [https://github.com/gemixin/digit-fsr-gui](https://github.com/gemixin/digit-fsr-gui).

The software simultaneously recorded tactile images and FSR measurements throughout data collection.

Each tactile interaction was performed using one of two exploratory motions:

- **Sliding:** controlled movement across an object surface while maintaining contact.
- **Rotation:** controlled rotational movement about a target contact point.

Every interaction consists of a sequence of 120 retained tactile frames collected from a labelled object region while maintaining a target force level. Every effort was made to ensure that contact remained within the target region throughout an interaction. However, for certain objects and regions, complete isolation was not always possible. For example, narrow football seams may include portions of adjacent panels due to their limited width.

Contact points were sampled within each labelled region to ensure diverse coverage of the object surface, with care taken to include any distinctive or unique features. 

To maintain balanced object representation, the number of interactions collected per region was adjusted according to the number of labelled regions on each object. Consequently, every training object contributes the same number of tactile interactions, preventing objects with more annotated regions from disproportionately influencing model training.

### Collection Counts

Each training object contributes 72 interactions: objects with two regions have 6 sliding and 6 rotation interactions per region per force level; objects with three regions have 4 of each motion per region per force level. Each object therefore contributes 8,640 frames.

In each non-training split, every region has 1 sliding and 1 rotation interaction per force level. A two-region object contributes 12 interactions (1,440 frames), while a three-region object contributes 18 interactions (2,160 frames).

The retained frames do not imply a fixed acquisition duration: recording performance, simultaneous force acquisition, and force-range filtering can affect the time required to collect 120 frames.

### Force Annotation

To improve consistency across interactions, FSR measurements were used to categorise contact into three predefined force levels. These force levels were designed to capture meaningful differences in sensor deformation while remaining practical for consistent human application.

| Force Level | Description  |
| ----------- | ------------ |
| 1       | Contact Only |
| 2       | Light Press  |
| 3       | Firm Press   |

During collection, a target force level was selected within the acquisition software. Real-time FSR measurements were used to monitor the applied force, while a connected LED bar provided visual feedback to the operator. The software additionally filtered tactile frames according to the selected target force level, ensuring that only samples falling within the predefined range for the chosen force category were retained.

FSR measurements were calibrated using a dedicated calibration framework: [https://github.com/gemixin/digit-force-calibration](https://github.com/gemixin/digit-force-calibration).

The dataset includes both raw FSR voltage measurements and approximate calibrated force estimates derived from the rear-mounted force sensor. While these values provide additional information about relative contact intensity, the primary force annotation is the discrete force-level label. The calibrated force estimates should therefore be considered approximate rather than laboratory-grade force ground truth.

### Semantic Annotations

The material labels, perceived-hardness labels, and natural-language descriptions were assigned by a human annotator and subsequently reviewed with the assistance of large language models to improve consistency. Descriptions are specific to each physical instance and region, and may differ between matched instances sharing the same category label.

---

## Dataset Structure

The Hugging Face version of Touch-Ex uses a flattened tabular representation where each row corresponds to a single visuo-tactile frame.

### Features

| Feature             | Type    | Description                                          |
| ------------------- | ------- | ---------------------------------------------------- |
| image           | Image   | DIGIT visuo-tactile image                            |
| object          | string  | Object-category label, shared by matched instances   |
| region          | string  | Region label                                         |
| object_region   | string  | Combined object-region label                         |
| force_level     | string  | Discrete force category (1, 2 or 3)                  |
| motion          | string  | Interaction motion type (sliding or rotation)        |
| fsr_voltage     | float32 | Raw force-sensitive resistor voltage                 |
| force_n         | float32 | Approximate calibrated force estimate (Newtons)      |
| hardness        | string  | Perceived hardness label                             |
| material        | string  | Primary material label                               |
| description     | string  | Natural-language tactile description                 |
| interaction_num | string  | Interaction number within object, region and force level |
| frame_num       | string  | Frame number within interaction                      |
| interaction_id  | string  | Unique interaction identifier		           |
### Splits
Touch-Ex is divided into four predefined Hugging Face dataset splits:
| Split | Physical objects | Purpose |
| --- | ---: | --- |
| `train` | 14 | Model training, validation, and in-distribution testing using interaction-level partitions |
| `validation_unseen_matched` | 4 | Model and configuration selection using new physical instances of known categories |
| `test_unseen_matched` | 4 | Final evaluation on new physical instances of known categories |
| `test_unseen_related` | 4 | Final analysis of transfer to unfamiliar, tactually related object categories |
For evaluation, all frames originating from the same interaction should remain within the same dataset partition. Because interactions contain highly correlated frame sequences, frame-level train/test splitting is not recommended. The `interaction_id` field is provided to support leakage-free splitting.

### Supplementary Files

The repository includes a reference image and separate folders containing interaction metadata and region-level annotations for each split.

| File | Description |
| --- | --- |
| `baseline.jpg` | Reference DIGIT image captured without object contact. May be used for background subtraction and preprocessing. |
| `metadata/{split_name}_metadata.csv` | Interaction-level metadata containing object, region, force level, motion type, and interaction numbers. |
| `descriptions/{split_name}_descriptions.csv` | Region-level semantic annotations including material, perceived hardness, and instance-specific descriptions. |

---

## Dataset Contents

### Training Objects (`train`)

|#|Object|Object Label|Region Labels|Specification|
|---|---|---|---|---|
|1|Toilet roll|toilet_roll|body, ends|New, full roll of unbranded, quilted white toilet paper|
|2|Pringles tube (empty)|tube_pringles|body, lid, base|Empty 185g Original Flavour Pringles tube with lid; freshness seal removed|
|3|Tin can (sealed)|tin_can|body, lid, base|Sealed 415g tin of Heinz Baked Beans|
|4|Tennis ball|tennis_ball|body, seam|Standard yellow Penn tennis ball|
|5|Scissors|scissors|handle, blades|Large pair of general-purpose scissors|
|6|Hammer|hammer|head, neck, handle|Hammer with rubber-dimpled handle|
|7|Toothbrush|toothbrush|handle, head|Generic adult plastic toothbrush|
|8|Mug|mug|body, handle, rim|Plain white ceramic mug|
|9|Plastic bottle (empty)|plastic_bottle|body, cap, base|Empty 50cl Nestlé Pure Life still water bottle with lid; label removed|
|10|Sponge scourer|sponge_scourer|sponge_side, scourer_side|Generic sponge scourer|
|11|TV remote|tv_remote|body, buttons|Splinktech universal Samsung TV remote control|
|12|Wooden spoon|wooden_spoon|handle, head|30cm wooden spoon|
|13|Tea towel (folded)|tea_towel|surface, edge|Cotton tea towel with lattice-effect|
|14|Football|football|panels, seams|Mitre Impel L30P size 3 football|
![Training objects](images/objects/train_objects.jpg)
### Matched Validation Objects (`validation_unseen_matched`)
| # | Object | Object Label | Region Labels | Specification |
| --- | --- | --- | --- | --- |
| 1 | Tin can (sealed) | tin_can | body, lid, base | Sealed 400g tin of Bramwells Vegetable Soup |
| 2 | Scissors | scissors | handle, blades | Medium pair of Scotch general-purpose scissors |
| 3 | Mug | mug | body, handle, rim | Patterned white ceramic mug |
| 4 | TV remote | tv_remote | body, buttons | Sony TV remote control |
### Matched Test Objects (`test_unseen_matched`)
| # | Object | Object Label | Region Labels | Specification |
| --- | --- | --- | --- | --- |
| 1 | Toothbrush | toothbrush | handle, head | Colgate Extra Clean Medium adult toothbrush |
| 2 | Plastic bottle (empty) | plastic_bottle | body, cap, base | Empty 500ml Juice Burst bottle with lid; label removed |
| 3 | Tea towel (folded) | tea_towel | surface, edge | Multi-coloured cotton tea towel with lattice-effect |
| 4 | Football | football | panels, seams | Adidas Starlancer Club size 5 football |
### Related Test Objects (`test_unseen_related`)
| # | Object | Object Label | Region Labels | Specification |
| --- | --- | --- | --- | --- |
| 1 | Table knife | table_knife | handle, blade | Standard stainless steel table knife |
| 2 | Dish brush | dish_brush | handle, head | Spontex pink washing-up brush |
| 3 | Beaker | beaker | body, rim | Patterned plastic beaker |
| 4 | Rubber ball | rubber_ball | body, ridges | Chuckit! Ultra medium rubber ball |

![Objects from the matched validation, matched test, and related test splits](images/objects/all_test_val_objects.jpg)

*The photograph shows the twelve physical objects listed across the three non-training splits above.*

---

## Usage

```python
from datasets import load_dataset
dataset = load_dataset("gemixin/touch-ex")
train = dataset["train"]
validation_unseen_matched = dataset["validation_unseen_matched"]
test_unseen_matched = dataset["test_unseen_matched"]
test_unseen_related = dataset["test_unseen_related"]
sample = train[0] 
image = sample["image"]
print({key: value for key, value in sample.items() if key != "image"})
```

The `image` field is automatically loaded as a PIL image object by the Hugging Face Datasets library.

---

## Supported Tasks

Touch-Ex supports a broad range of visuo-tactile learning tasks, including:

- Object classification
- Object-region classification
- Force-level classification
- Force-aware tactile perception
- Sequential tactile modelling
- Representation learning
- Tactile-language alignment
- Multimodal learning
- Active tactile exploration

Reference implementations, baseline models, and experiments are available in the Touch-Ex repository: [https://github.com/gemixin/touch-ex](https://github.com/gemixin/touch-ex).

---

## Dataset Statistics

| Split | Physical objects | Object regions | Interactions | Frames |
| --- | ---: | ---: | ---: | ---: |
| `train` | 14 | 33 | 1,008 | 120,960 |
| `validation_unseen_matched` | 4 | 10 | 60 | 7,200 |
| `test_unseen_matched` | 4 | 9 | 54 | 6,480 |
| `test_unseen_related` | 4 | 8 | 48 | 5,760 |
| **Total** | **26** | **60** | **1,170** | **140,400** |

Region counts refer to regions of physical instances, so matched instances are counted separately. The dataset spans 18 distinct object categories, three force levels, and two motion types, with 120 frames per interaction.

---

## Related Resources

### Dataset Collection Software

[https://github.com/gemixin/digit-fsr-gui](https://github.com/gemixin/digit-fsr-gui)

### Force Calibration Framework

[https://github.com/gemixin/digit-force-calibration](https://github.com/gemixin/digit-force-calibration)

### Touch-Ex Models and Experiments

[https://github.com/gemixin/touch-ex](https://github.com/gemixin/touch-ex)

### Raw Dataset

The original hierarchical version of Touch-Ex, including the complete folder structure, interaction-level metadata, semantic annotation files, and baseline image, is available here:

[https://drive.google.com/drive/folders/1Jw3O8ewLCXLjAZWtHekXgwb4KDu-blX2?usp=sharing](https://drive.google.com/drive/folders/1Jw3O8ewLCXLjAZWtHekXgwb4KDu-blX2?usp=sharing)

---

## Limitations

- Data was collected using a single DIGIT tactile sensor.
- Interactions were performed manually by a single operator.
- Force measurements are intended primarily to support force-level categorisation and should not be interpreted as highly accurate force ground truth.
- Objects are limited to household items collected in the United Kingdom.
- The dataset does not include external RGB images or video streams of the objects.
- Data was collected in a home office environment rather than a controlled laboratory setting.
- Environmental conditions were not explicitly varied during collection.

---

## Ethical Considerations

The dataset contains visuo-tactile observations of inanimate household objects and does not contain personal information, biometric data, human subjects, or sensitive content.

The dataset is intended for research in tactile perception, robotics, multimodal learning, and representation learning.

---

## License

Touch-Ex is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence.

You are free to:

- Share: copy and redistribute the dataset in any medium or format.
- Adapt: remix, transform, and build upon the dataset for any purpose, including commercial use.

Under the following condition:

- Attribution: appropriate credit must be given to the dataset authors and any changes made must be indicated.

For the full licence text, see:

https://creativecommons.org/licenses/by/4.0/

---

## Citation

If you use Touch-Ex in your research, please cite:

```bibtex
@dataset{mclean2026touchex,
  author = {McLean, Gemma and Hao, Zhou Daniel},
  title = {Touch-Ex: A Region-Level, Force-Annotated Visuo-Tactile Dataset},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/gemixin/touch-ex}
}
```

Touch-Ex was collected using the DIGIT vision-based tactile sensor. If you use this dataset in your research, please additionally cite the original DIGIT paper:

```bibtex
@article{lambeta2020digit,
  title = {DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor with Application to In-Hand Manipulation},
  author = {Lambeta, Mike and Chou, Po-Wei and Tian, Stephen and Yang, Brian and Maloon, Benjamin and Most, Victoria Rose and Stroud, Dave and Santos, Raymond and Byagowi, Ahmad and Kammerer, Gregg and Jayaraman, Dinesh and Calandra, Roberto},
  journal = {IEEE Robotics and Automation Letters},
  volume = {5},
  number = {3},
  pages = {3838--3845},
  year = {2020},
  doi = {10.1109/LRA.2020.2977257}
}
```

	---
	language:
	- en
	license: cc-by-4.0

	pretty_name: Touch-Ex

	task_categories:
	- image-classification
	- feature-extraction

	tags:
	- tactile-sensing
	- visuo-tactile
	- robotics
	- digit-sensor
	- multimodal
	- tactile-language
	- tactile-perception
	- representation-learning
	- force-aware-perception
	---

	# Touch-Ex: A Region-Level, Force-Annotated Visuo-Tactile Dataset

	Developed by Gemma McLean and supervised by Dr Daniel Hao, University of Leicester.

	## Dataset Summary

	Touch-Ex (Touch Exploration) is a region-level, force-annotated visuo-tactile dataset collected using a DIGIT vision-based tactile sensor and a collection of common UK household objects. The dataset was designed to support research in tactile representation learning, force-aware perception, object-region recognition, temporal tactile modelling, tactile-language alignment and in particular, active tactile exploration.

	The dataset contains 140,400 visuo-tactile frames collected across 1,170 tactile interactions from 26 physical objects spanning 18 object categories. The objects cover varied materials, textures, geometries, and compliance levels.

	Touch-Ex is divided into four predefined splits:

	\| Split \| Physical objects \| Purpose \|
	\| --- \| ---: \| --- \|
	\| `train` \| 14 \| Model training, validation, and in-distribution testing using interaction-level partitions \|
	\| `validation_unseen_matched` \| 4 \| Model and configuration selection using new physical instances of known categories \|
	\| `test_unseen_matched` \| 4 \| Final evaluation on new physical instances of known categories \|
	\| `test_unseen_related` \| 4 \| Final analysis of transfer to unfamiliar, tactually related object categories \|

	Each dataset sample corresponds to a single tactile frame and includes image data, force measurements, interaction metadata, semantic descriptions, and structured tactile attributes.

	The training split was designed so that each object contributes an equal number of tactile interactions regardless of the number of annotated regions, while other splits are intentionally smaller.

	## Example Samples

	\| image \| object \| region \| force_level \| motion \| material \| hardness \| description \|
	\|--------------\|---------\|---------\|---------\|---------\|---------\|---------\|---------\|
	\| ![](images/examples/hammer_handle_sliding_fl3.jpg) \| hammer \| handle \| 3 \| sliding \| rubber \| medium-hard \| An elongated, solid rubber-coated gripping area featuring small, evenly spaced circular dimples that provide a repeating texture. \|
	\| ![](images/examples/scissors_blades_rotation_fl2.jpg) \| scissors \| blades \| 2 \| rotation \| metal \| hard \| A pair of elongated, flat metallic blades with polished surfaces that gradually taper to sharp outer edges and are joined at a raised pivot screw. \|
	\| ![](images/examples/tv_remote_buttons_sliding_fl1.jpg) \| tv_remote \| buttons \| 1 \| sliding \| rubber \| soft-medium \| Small, compact raised rubber features, including rectangles, squares, and circles, that provide springy resistance when pressed. \|

	Touch-Ex samples showing tactile images and their key annotations.

	---

	## Dataset Modality

	Following standard terminology for optical tactile sensors, "visuo-tactile" refers strictly to the internal image data captured by the DIGIT sensor's embedded camera. This dataset intentionally excludes external RGB camera views of the objects to prioritise compactness and pure tactile-language representation learning.

	---

	## Why Touch-Ex?

	Touch-Ex was designed to address several limitations of existing visuo-tactile datasets. While many tactile datasets focus primarily on object identity or contact localisation, Touch-Ex introduces region-level annotations, force-level annotations, and semantic tactile descriptions within a unified benchmark.

	Touch-Ex combines:
	- Region-level object annotations
	- Force-level tactile annotations
	- Sequential tactile interactions
	- Controlled exploratory motions
	- Natural-language tactile descriptions
	- Structured semantic attributes

	In addition to standard in-distribution evaluation, Touch-Ex separates generalisation validation from final testing. Matched-instance validation supports model and configuration selection, while separate matched-instance and related-category test splits support final assessment of transfer beyond the training objects. This enables systematic evaluation of object-level generalisation and the extent to which learned representations capture transferable tactile properties rather than object identity alone.

	---

	## Data Collection

	Touch-Ex was collected using a DIGIT vision-based tactile sensor operated manually by a human data collector, moving and orienting the DIGIT and the object as needed.

	To enable force-aware data collection, a Grove Round Force Sensor (FSR402) was attached to the rear of the DIGIT sensor and connected to a Raspberry Pi Pico via a Grove Shield.

	During data collection, the operator applied pressure to the rear of the sensor using their index finger (stabilising either side with their thumb and middle finger) while bringing the DIGIT into contact with the target object. The rear-mounted FSR measured this applied force, providing an estimate of how strongly the sensor was pressed against the object surface.

	Data acquisition was performed using a custom collection tool developed specifically for Touch-Ex: [https://github.com/gemixin/digit-fsr-gui](https://github.com/gemixin/digit-fsr-gui).

	The software simultaneously recorded tactile images and FSR measurements throughout data collection.

	Each tactile interaction was performed using one of two exploratory motions:

	- Sliding: controlled movement across an object surface while maintaining contact.
	- Rotation: controlled rotational movement about a target contact point.

	Every interaction consists of a sequence of 120 retained tactile frames collected from a labelled object region while maintaining a target force level. Every effort was made to ensure that contact remained within the target region throughout an interaction. However, for certain objects and regions, complete isolation was not always possible. For example, narrow football seams may include portions of adjacent panels due to their limited width.

	Contact points were sampled within each labelled region to ensure diverse coverage of the object surface, with care taken to include any distinctive or unique features.

	To maintain balanced object representation, the number of interactions collected per region was adjusted according to the number of labelled regions on each object. Consequently, every training object contributes the same number of tactile interactions, preventing objects with more annotated regions from disproportionately influencing model training.

	### Collection Counts

	Each training object contributes 72 interactions: objects with two regions have 6 sliding and 6 rotation interactions per region per force level; objects with three regions have 4 of each motion per region per force level. Each object therefore contributes 8,640 frames.

	In each non-training split, every region has 1 sliding and 1 rotation interaction per force level. A two-region object contributes 12 interactions (1,440 frames), while a three-region object contributes 18 interactions (2,160 frames).

	The retained frames do not imply a fixed acquisition duration: recording performance, simultaneous force acquisition, and force-range filtering can affect the time required to collect 120 frames.

	### Force Annotation

	To improve consistency across interactions, FSR measurements were used to categorise contact into three predefined force levels. These force levels were designed to capture meaningful differences in sensor deformation while remaining practical for consistent human application.

	\| Force Level \| Description \|
	\| ----------- \| ------------ \|
	\| 1 \| Contact Only \|
	\| 2 \| Light Press \|
	\| 3 \| Firm Press \|

	During collection, a target force level was selected within the acquisition software. Real-time FSR measurements were used to monitor the applied force, while a connected LED bar provided visual feedback to the operator. The software additionally filtered tactile frames according to the selected target force level, ensuring that only samples falling within the predefined range for the chosen force category were retained.

	FSR measurements were calibrated using a dedicated calibration framework: [https://github.com/gemixin/digit-force-calibration](https://github.com/gemixin/digit-force-calibration).

	The dataset includes both raw FSR voltage measurements and approximate calibrated force estimates derived from the rear-mounted force sensor. While these values provide additional information about relative contact intensity, the primary force annotation is the discrete force-level label. The calibrated force estimates should therefore be considered approximate rather than laboratory-grade force ground truth.

	### Semantic Annotations

	The material labels, perceived-hardness labels, and natural-language descriptions were assigned by a human annotator and subsequently reviewed with the assistance of large language models to improve consistency. Descriptions are specific to each physical instance and region, and may differ between matched instances sharing the same category label.

	---

	## Dataset Structure

	The Hugging Face version of Touch-Ex uses a flattened tabular representation where each row corresponds to a single visuo-tactile frame.

	### Features

	\| Feature \| Type \| Description \|
	\| ------------------- \| ------- \| ---------------------------------------------------- \|
	\| image \| Image \| DIGIT visuo-tactile image \|
	\| object \| string \| Object-category label, shared by matched instances \|
	\| region \| string \| Region label \|
	\| object_region \| string \| Combined object-region label \|
	\| force_level \| string \| Discrete force category (1, 2 or 3) \|
	\| motion \| string \| Interaction motion type (sliding or rotation) \|
	\| fsr_voltage \| float32 \| Raw force-sensitive resistor voltage \|
	\| force_n \| float32 \| Approximate calibrated force estimate (Newtons) \|
	\| hardness \| string \| Perceived hardness label \|
	\| material \| string \| Primary material label \|
	\| description \| string \| Natural-language tactile description \|
	\| interaction_num \| string \| Interaction number within object, region and force level \|
	\| frame_num \| string \| Frame number within interaction \|
	\| interaction_id \| string \| Unique interaction identifier \|

	### Splits

	Touch-Ex is divided into four predefined Hugging Face dataset splits:

	\| Split \| Physical objects \| Purpose \|
	\| --- \| ---: \| --- \|
	\| `train` \| 14 \| Model training, validation, and in-distribution testing using interaction-level partitions \|
	\| `validation_unseen_matched` \| 4 \| Model and configuration selection using new physical instances of known categories \|
	\| `test_unseen_matched` \| 4 \| Final evaluation on new physical instances of known categories \|
	\| `test_unseen_related` \| 4 \| Final analysis of transfer to unfamiliar, tactually related object categories \|

	For evaluation, all frames originating from the same interaction should remain within the same dataset partition. Because interactions contain highly correlated frame sequences, frame-level train/test splitting is not recommended. The `interaction_id` field is provided to support leakage-free splitting.

	### Supplementary Files

	The repository includes a reference image and separate folders containing interaction metadata and region-level annotations for each split.

	\| File \| Description \|
	\| --- \| --- \|
	\| `baseline.jpg` \| Reference DIGIT image captured without object contact. May be used for background subtraction and preprocessing. \|
	\| `metadata/{split_name}_metadata.csv` \| Interaction-level metadata containing object, region, force level, motion type, and interaction numbers. \|
	\| `descriptions/{split_name}_descriptions.csv` \| Region-level semantic annotations including material, perceived hardness, and instance-specific descriptions. \|

	---

	## Dataset Contents

	### Training Objects (`train`)

	\|#\|Object\|Object Label\|Region Labels\|Specification\|
	\|---\|---\|---\|---\|---\|
	\|1\|Toilet roll\|toilet_roll\|body, ends\|New, full roll of unbranded, quilted white toilet paper\|
	\|2\|Pringles tube (empty)\|tube_pringles\|body, lid, base\|Empty 185g Original Flavour Pringles tube with lid; freshness seal removed\|
	\|3\|Tin can (sealed)\|tin_can\|body, lid, base\|Sealed 415g tin of Heinz Baked Beans\|
	\|4\|Tennis ball\|tennis_ball\|body, seam\|Standard yellow Penn tennis ball\|
	\|5\|Scissors\|scissors\|handle, blades\|Large pair of general-purpose scissors\|
	\|6\|Hammer\|hammer\|head, neck, handle\|Hammer with rubber-dimpled handle\|
	\|7\|Toothbrush\|toothbrush\|handle, head\|Generic adult plastic toothbrush\|
	\|8\|Mug\|mug\|body, handle, rim\|Plain white ceramic mug\|
	\|9\|Plastic bottle (empty)\|plastic_bottle\|body, cap, base\|Empty 50cl Nestlé Pure Life still water bottle with lid; label removed\|
	\|10\|Sponge scourer\|sponge_scourer\|sponge_side, scourer_side\|Generic sponge scourer\|
	\|11\|TV remote\|tv_remote\|body, buttons\|Splinktech universal Samsung TV remote control\|
	\|12\|Wooden spoon\|wooden_spoon\|handle, head\|30cm wooden spoon\|
	\|13\|Tea towel (folded)\|tea_towel\|surface, edge\|Cotton tea towel with lattice-effect\|
	\|14\|Football\|football\|panels, seams\|Mitre Impel L30P size 3 football\|

	![Training objects](images/objects/train_objects.jpg)

	### Matched Validation Objects (`validation_unseen_matched`)

	\| # \| Object \| Object Label \| Region Labels \| Specification \|
	\| --- \| --- \| --- \| --- \| --- \|
	\| 1 \| Tin can (sealed) \| tin_can \| body, lid, base \| Sealed 400g tin of Bramwells Vegetable Soup \|
	\| 2 \| Scissors \| scissors \| handle, blades \| Medium pair of Scotch general-purpose scissors \|
	\| 3 \| Mug \| mug \| body, handle, rim \| Patterned white ceramic mug \|
	\| 4 \| TV remote \| tv_remote \| body, buttons \| Sony TV remote control \|

	### Matched Test Objects (`test_unseen_matched`)

	\| # \| Object \| Object Label \| Region Labels \| Specification \|
	\| --- \| --- \| --- \| --- \| --- \|
	\| 1 \| Toothbrush \| toothbrush \| handle, head \| Colgate Extra Clean Medium adult toothbrush \|
	\| 2 \| Plastic bottle (empty) \| plastic_bottle \| body, cap, base \| Empty 500ml Juice Burst bottle with lid; label removed \|
	\| 3 \| Tea towel (folded) \| tea_towel \| surface, edge \| Multi-coloured cotton tea towel with lattice-effect \|
	\| 4 \| Football \| football \| panels, seams \| Adidas Starlancer Club size 5 football \|

	### Related Test Objects (`test_unseen_related`)

	\| # \| Object \| Object Label \| Region Labels \| Specification \|
	\| --- \| --- \| --- \| --- \| --- \|
	\| 1 \| Table knife \| table_knife \| handle, blade \| Standard stainless steel table knife \|
	\| 2 \| Dish brush \| dish_brush \| handle, head \| Spontex pink washing-up brush \|
	\| 3 \| Beaker \| beaker \| body, rim \| Patterned plastic beaker \|
	\| 4 \| Rubber ball \| rubber_ball \| body, ridges \| Chuckit! Ultra medium rubber ball \|

	![Objects from the matched validation, matched test, and related test splits](images/objects/all_test_val_objects.jpg)

	The photograph shows the twelve physical objects listed across the three non-training splits above.

	---

	## Usage

	```python
	from datasets import load_dataset

	dataset = load_dataset("gemixin/touch-ex")

	train = dataset["train"]
	validation_unseen_matched = dataset["validation_unseen_matched"]
	test_unseen_matched = dataset["test_unseen_matched"]
	test_unseen_related = dataset["test_unseen_related"]

	sample = train[0]
	image = sample["image"]
	print({key: value for key, value in sample.items() if key != "image"})
	```

	The `image` field is automatically loaded as a PIL image object by the Hugging Face Datasets library.

	---

	## Supported Tasks

	Touch-Ex supports a broad range of visuo-tactile learning tasks, including:

	- Object classification
	- Object-region classification
	- Force-level classification
	- Force-aware tactile perception
	- Sequential tactile modelling
	- Representation learning
	- Tactile-language alignment
	- Multimodal learning
	- Active tactile exploration

	Reference implementations, baseline models, and experiments are available in the Touch-Ex repository: [https://github.com/gemixin/touch-ex](https://github.com/gemixin/touch-ex).

	---

	## Dataset Statistics

	\| Split \| Physical objects \| Object regions \| Interactions \| Frames \|
	\| --- \| ---: \| ---: \| ---: \| ---: \|
	\| `train` \| 14 \| 33 \| 1,008 \| 120,960 \|
	\| `validation_unseen_matched` \| 4 \| 10 \| 60 \| 7,200 \|
	\| `test_unseen_matched` \| 4 \| 9 \| 54 \| 6,480 \|
	\| `test_unseen_related` \| 4 \| 8 \| 48 \| 5,760 \|
	\| Total \| 26 \| 60 \| 1,170 \| 140,400 \|

	Region counts refer to regions of physical instances, so matched instances are counted separately. The dataset spans 18 distinct object categories, three force levels, and two motion types, with 120 frames per interaction.

	---

	## Related Resources

	### Dataset Collection Software

	[https://github.com/gemixin/digit-fsr-gui](https://github.com/gemixin/digit-fsr-gui)

	### Force Calibration Framework

	[https://github.com/gemixin/digit-force-calibration](https://github.com/gemixin/digit-force-calibration)

	### Touch-Ex Models and Experiments

	[https://github.com/gemixin/touch-ex](https://github.com/gemixin/touch-ex)

	### Raw Dataset

	The original hierarchical version of Touch-Ex, including the complete folder structure, interaction-level metadata, semantic annotation files, and baseline image, is available here:

	[https://drive.google.com/drive/folders/1Jw3O8ewLCXLjAZWtHekXgwb4KDu-blX2?usp=sharing](https://drive.google.com/drive/folders/1Jw3O8ewLCXLjAZWtHekXgwb4KDu-blX2?usp=sharing)

	---

	## Limitations

	- Data was collected using a single DIGIT tactile sensor.
	- Interactions were performed manually by a single operator.
	- Force measurements are intended primarily to support force-level categorisation and should not be interpreted as highly accurate force ground truth.
	- Objects are limited to household items collected in the United Kingdom.
	- The dataset does not include external RGB images or video streams of the objects.
	- Data was collected in a home office environment rather than a controlled laboratory setting.
	- Environmental conditions were not explicitly varied during collection.

	---

	## Ethical Considerations

	The dataset contains visuo-tactile observations of inanimate household objects and does not contain personal information, biometric data, human subjects, or sensitive content.

	The dataset is intended for research in tactile perception, robotics, multimodal learning, and representation learning.

	---

	## License

	Touch-Ex is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence.

	You are free to:

	- Share: copy and redistribute the dataset in any medium or format.
	- Adapt: remix, transform, and build upon the dataset for any purpose, including commercial use.

	Under the following condition:

	- Attribution: appropriate credit must be given to the dataset authors and any changes made must be indicated.

	For the full licence text, see:

	https://creativecommons.org/licenses/by/4.0/

	---

	## Citation

	If you use Touch-Ex in your research, please cite:

	```bibtex
	@dataset{mclean2026touchex,
	author = {McLean, Gemma and Hao, Zhou Daniel},
	title = {Touch-Ex: A Region-Level, Force-Annotated Visuo-Tactile Dataset},
	year = {2026},
	publisher = {Hugging Face},
	url = {https://huggingface.co/datasets/gemixin/touch-ex}
	}
	```

	Touch-Ex was collected using the DIGIT vision-based tactile sensor. If you use this dataset in your research, please additionally cite the original DIGIT paper:

	```bibtex
	@article{lambeta2020digit,
	title = {DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor with Application to In-Hand Manipulation},
	author = {Lambeta, Mike and Chou, Po-Wei and Tian, Stephen and Yang, Brian and Maloon, Benjamin and Most, Victoria Rose and Stroud, Dave and Santos, Raymond and Byagowi, Ahmad and Kammerer, Gregg and Jayaraman, Dinesh and Calandra, Roberto},
	journal = {IEEE Robotics and Automation Letters},
	volume = {5},
	number = {3},
	pages = {3838--3845},
	year = {2020},
	doi = {10.1109/LRA.2020.2977257}
	}
	```