Datasets

Artificial Intelligence

AI Datasets

Qualcomm Technologies, Inc. has published a variety of datasets for research use. We’ve brought together datasets you can use to train models in the kinds of applications most common to mobile computing, including these:

Advanced Driver Assistance Systems (ADAS)
Extended Reality (XR): VR and AR applications
Smartphones
Robotics
Smart homes
Security
Industrial IoT
Healthcare and assistive technology
Keyword spotting
Tensor program execution times

Whether your applications depend on recognizing gestures, speech or images, you’ll find datasets you can use for training in machine learning and artificial intelligence.

Jester

The Jester gesture recognition dataset comprises close to 150k labeled video clips demonstrating common hand gestures.

Something-Something v. 2

This model recognizes certain simple, single-frame gestures utilizing common everyday objects.

CasualCircuit

The dataset showcases images depicting a robotic arm engaging with a button and light system.

Wireless Indoor Simulations

The dataset contains the geometry of the propagation environment and the wireless channels.

Massive MIMO Spatial Channel
Model Dataset

The data can be used to better understand the statistical distribution of channel characteristics in a typical dense urban layout.

Qualcomm Keyword Speech Dataset

Keyword spotting (KWS) is widely used for detecting specific keywords in personal devices like mobile phones and home appliances.

Qualcomm Dataset of Tensor Programs
Execution Times – QAST

This dataset is intended for research purposes only and to support and contribute to the graph research community.

Qualcomm® Rasterized Images

The Qualcomm Rasterized Images for Super-resolution Processing dataset was created to facilitate the development and research of super-resolution algorithms for gaming.

ClevrSkills Dataset

ClevrSkills includes 330,000 robot episodes/trajectories including videos (from multiple views), corresponding actions, and other annotations including text, bounding boxes, cameras poses etc., which were generated from over 33 tasks in ClevrSkills environment suite.

Qualcomm Exercise Video Dataset (QEVD)

Qualcomm AI Research presents the Qualcomm Exercise Video Dataset (QEVD) which explores human-AI interaction in the challenging real-world domain of fitness coaching.

PlotTwist Dataset

This dataset is suitable for visual reasoning with Vision-Language Models (VLMs) to answer human-readable questions about images of mathematical graphs.

AirLetters Dataset

AirLetters dataset, with over 161,000 video-label pairs of video clips, is a benchmark for evaluating a model’s ability to classify articulated motions correctly.

Qualcomm Interactive Video Dataset (QIVD)

QIVD is a benchmark and dataset designed to improve a model's ability to answer questions about events occurring in videos.