Table 1.
Summary of datasets included in this study.
| Population | Investigated activities | Measurement parameters | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Dataset name | Dataset acronym | N | Sex (male) | Age (y) | Height (cm) | Weight (kg) | BMI (kg/m2) | Condition | Sensing device | Approximate sensor location | MR (g) | AR (bit) | SR (Hz) | Ref. | |
| WISDMs Actitracker activity prediction dataset v2.0 | Actitracker | 166* | 15 (6 females) | 19–51 (30.5 ± 10.8) | 163–188 (174.5 ± 6.8) | 51–109 (75.8 ± 16.6) | 19–35 (24.8 ± 4.6) | Normal walking, sitting, standing, lying, running | Free-living | Smartphone: Android-based | Unspecified | N/A | N/A | 20 | 41 |
| Complex Human Activity Dataset | CHA | 10** | 10 | 23–35 | N/A | N/A | N/A | Normal walking, ascending stairs, descending stairs, sitting, standing, typing, handwriting, eating, drinking, jogging, cycling, giving a talk, smoking | Controlled | Smartphone: Samsung Galaxy S2 | Thigh and wrist | N/A | N/A | 50 | 42 |
| Daily Life Activities Dataset | DaLiAc | 19 | 11 | 18–55 (26.5 ± 7.7) | 158–196 (177.0 ± 11.1) | 54–108 (75.2 ± 14.2) | 17–34 (23.9 ± 3.7) | Normal walking, ascending stairs, descending stairs, lying, sitting, standing, washing dishes, vacuuming, sweeping, running, cycling (50 W and 100 W), rope jumping | Controlled | Wearable accelerometer: SHIMMER | Waist, chest, and wrist | ±6 | 12 | 204.8 | 43 |
| Dataset for behavioral context recognition in-the-wild from mobile sensor | Extrasensory | 60** | 26 | 18–42 (24 ± 5) |
145–188 (171 ± 9) |
50–93 (66 ± 11) |
18–32 (23 ± 3) |
Normal walking, ascending stairs, descending stairs, sitting, standing, lying, watching TV, handwriting, eating, using motorized transportation (car, bus, motor, train), cooking, washing dishes, dressing, grooming, sweeping, running, cycling, jumping, skateboarding | Free-living | Smartphone: Android- and iOS-based | Unspecified | N/A | N/A | 33 | 44 |
| Human Activities and Postural Transitions Dataset | HAPT | 30** | N/A | 19–48 | N/A | N/A | N/A | Normal walking, ascending stairs, descending stairs, standing, sitting, lying, body transitions (standing to sitting, sitting to standing, sitting to lying, lying to sitting, standing to lying, and lying to standing) | Controlled | Smartphone: Samsung Galaxy S2 | Waist | N/A | N/A | 50 | 45 |
| Human Activity Sensing Consortium – Pedestrian Activity Corpus 2016 | HASC | 539 | 438 | 15–69 (28.6 ± 12.2) | 147–189 (169.4 ± 7.9) | 37–118 (62.8 ± 11.5) | 15–38 (21.8 ± 3.4) | Normal walking, ascending stairs, descending stairs, standing, jogging, jumping | Controlled, Free-living | Smartphone: Android- and iOS-based | Thigh, waist, chest, arm, wrist, and unspecified | N/A | N/A | 100 | 46 |
| Public Dataset of Accelerometer Data for Human Motion Primitives Detection | HMPD | 16** | 11 | 19–81 (57.4) | N/A | 56–85 (72.7) | N/A | Normal walking, ascending stairs, descending stairs, drinking, pouring, eating soup or meat, combing hair, brushing teeth, using telephone, body transitions (standing to lying, lying to standing, standing to sitting, and sitting to standing) | Controlled | Wearable accelerometer | Wrist | ±1.5 | 6 | 32 | 47 |
| Identification of Walking, Stair Climbing, and Driving using Wearable Accelerometers | IWSCD | 32 | 13 | 23–52 (39.0 ± 9.0) | 147–193 (173.5 ± 11.1) | 45–140 (77.0 ± 22.9) | 18–40 (25.2 ± 5.6) | Normal walking, ascending stairs, descending stairs, using motorized transportation (car) | Controlled | Wearable accelerometer: ActiGraph GT3X + | Waist and wrist | N/A | N/A | 100 | 48 |
| Mobile Health Dataset | MHEALTH | 10** | N/A | N/A | N/A | N/A | N/A | Normal walking, lying, sitting, standing, cycling, jogging, running, forward and backward jumping, body stretching (bending waist forward, elevating arm, crouching) | Controlled | Wearable accelerometer: SHIMMER2 | Chest and wrist | N/A | N/A | 50 | 49 |
| Recognition of Activities of Daily Living using Smartphones | MobiAct | 61 | 42 | 20–40 (24.9 ± 3.7) | 158–193 (175.9 ± 8.1) | 50–120 (76.8 ± 15.0) | 18–35 (24.7 ± 3.8) | Normal walking, ascending stairs, descending stairs, lying, sitting, standing, jogging, jumping, body transitions (standing to sitting [on a chair, in a car], sitting to standing [from a chair, from a car]) | Controlled | Smartphone: Samsung Galaxy S3 | Thigh | N/A | N/A | 200 | 50 |
| Sensor Based Human Activity and Attribute Recognition | MotionSense | 24 | 14 | 18–46 (28.8 ± 5.4) | 161–190 (174.2 ± 8.9) | 48–102 (72.1 ± 16.2) | 18–32 (23.6 ± 4.1) | Normal walking, ascending stairs, descending stairs, sitting, standing, jogging | Controlled | Smartphone: iPhone 6 | Thigh | N/A | N/A | 50 | 51 |
| Physical Activity Recognition Dataset Using Smartphone Sensors | PARDUSS | 10** | 10 | 25–30 | N/A | N/A | N/A | Normal walking, ascending stairs, descending stairs, sitting, standing, jogging, cycling | Controlled | Smartphone: Samsung Galaxy S2 | Thigh, waist, arm, and wrist | N/A | N/A | 50 | 52 |
| Pedometer Evaluation Project | Pedometer | 30 | 15 | 19–27 (21.9 ± 52.4) | 152–193 (171.0 ± 10.8) | 43–136 (70.5 ± 17.6) | 17–37 (23.8 ± 3.7) | Normal walking | Controlled | Wearable sensor: SHIMMER3 | Waist and wrist | ±4 | N/A | 15 | 53 |
| Real-World Dataset | RealWorld | 15 | 8 | 16–62 (31.9 ± 12.4) | 163–183 (173.1 ± 6.9) | 48–95 (74.1 ± 13.8) | 18–35 (24.7 ± 4.4) | Normal walking, ascending stairs, descending stairs, lying, sitting, standing, running, jumping | Controlled | Smartphone: Samsung Galaxy S4, smartwatch: LG G Watch R | Smartphone: thigh, waist, chest, and arm, smartwatch: wrist | N/A | N/A | 50 | 32 |
| Speed-Breaker Dataset | SpeedBreaker | 40** | N/A | N/A | N/A | N/A | N/A | Using motorized transportation (car, motorcycle, and rickshaw) | Free-living | Smartphone: Android-based | Unspecified | N/A | N/A | 100 | 54 |
| Simulated Falls and Daily Living Activities Dataset | SFDLA | 17 | 10 | 19–27 (21.9 ± 2.0) | 157–184 (171.6 ± 7.8) | 47–92 (65.0 ± 13.9) | 17–31 (21.9 ± 3.7) | Normal walking, walking backwards, limping, jogging, squatting, bending, body transitions (lying to sitting, lying to standing, and standing to sitting) [on a chair, a sofa, a bed, in the air], coughing/sneezing | Controlled | Wearable accelerometer: Xsens MTw | Thigh, waist, chest, and wrist | ±12 | N/A | 25 | 55 |
| A Fall and Movement Dataset | SisFall | 38 | 19 | 19–75 (40.2 ± 21.3) | 149–183 (164.1 ± 9.3) | 41–102 (62.2 ± 12.6) | 18–35 (23.0 ± 3.5) | Normal walking (slow, fast), jogging (slow and fast), jumping, body transitions (rolling while lying, standing to sitting to standing [with a low and a high chair, in a car, slow and fast], sitting to lying to sitting [slow and fast], and sitting to standing to sitting) | Controlled | Wearable accelerometer: self-developed | Waist | ±16 | 13 | 200 | 56 |
| Human physical activity dataset | SPADES | 42 | 27 |
18–30 (23 ± 3) |
151–180 174 ± 8 |
51–112 (73 ± 15) |
18–35 (24 ± 4) |
Normal walking, ascending stairs, descending stairs, treadmill walk (1, 2, 3, and 3.5 mph), lying, sitting, standing, reclining, handwriting, typing, folding towels, filling shelves, sweeping, running, cycling, jumping jacks | Controlled | Wearable accelerometer: ActiGraph GT9X | Thigh, waist, and wrist | 8 | N/A | 80 | 57 |
| University of Milano Bicocca Smartphone-based Human Activity Recognition Dataset | UniMiB-SHAR | 30 | 6 | 18–60 (26.6 ± 11.6) | 160–190 (168.8 ± 6.8) | 50–82 (64.4 ± 9.8) | 18–27 (22.5 ± 2.5) | Normal walking, ascending stairs, descending stairs, running, jumping, body transitions (lying to standing, sitting to standing, standing to sitting, and standing to lying) | Controlled | Smartphone: Samsung Galaxy Nexus | Thigh | ±2 | 9 | 50 | 58 |
| Wireless Sensor Data Mining Dataset | WISDM | 51** | N/A | 18–25 | N/A | N/A | N/A | Normal walking, sitting, standing, jogging, eating soup, pasta, and chips, drinking, handwriting, typing, folding clothes, brushing teeth, clapping | Controlled | Smartphone: Google Nexus 5/5X and Samsung Galaxy S5, smartwatch: LG watch |
Smartphone: thigh, smartwatch: wrist |
N/A | N/A | 20 | 59 |
Age, height, weight, and BMI are provided as range (mean ± SD), when available.
Note: * detailed demographics available for some subjects; ** detailed demographics unavailable; BMI body mass index, MR measurement range, AR amplitude resolution, SR approximate sampling rate.