On Indoor Human Sensing Using Commodity Radar

Alloulah, Mohammed; Isopoussu, Anton; Kawsar, Fahim

doi:10.1145/3267305.3274180

Cited by 4 publications

(14 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In short, radar gesture performance studies deal little or nothing with user preferences, and preference studies deal little or nothing with performance. This paper reconciles performance with preference in the context of gestures recognized by a commodity radar [5] through multiple materials [32]. Nielsen [33] shed light on the debate between favoring preference and favoring performance, pointing out that the former is typically reported by subjective measures, while the latter is reported by objective measures.…”

Section: Introductionmentioning

confidence: 71%

“…Regarding GES1, the agreement rates are overall moderate in magnitude (M =0.24, SD=0.157), ranging between 0.074 (low) for ''Turn assistant help off'' and 0.536 (very high) for ''Answer help phone Call''. On the global sampling, 5 19 =26% of the rates belong to the low consensus category, 9 19 =47% of the rates belong to the moderate range, 3 19 =16% are high, and 2 19 =10% are very high. Apart from a few exceptions, most gestures received an agreement rate slightly higher or close to those reported in the GES literature ( [76] that summarized agreement rates of 18 GESs).…”

Section: ) Agreement Ratementioning

confidence: 99%

“…Yeo et al [47] used radar in tangible interaction for counting, ordering, and identifying objects for tracking their orientation, movement, and between-object distance. Radars are widely used in several domains of application, such as indoor human sensing with commodity radar [5], human activity recognition [8], human position estimation [38], motion classification [14].…”

Section: A Radar-based Gesture Interaction 1) Performancementioning

confidence: 99%

“…In view of all this work, we can apply a DL algorithm, but we are keen to put a template-based [50] recognizer [51] to the test because it allows us to customize the gestures by modifying templates only, with a reasonable number of templates [12], and without re-training [52], [53]. Furthermore, we are siding with the mobile radars that are widely available on the market by taking the Walabot DIY 2 radar [3], [5], [15], [26], [45], [54], [55].…”

Section: A Radar-based Gesture Interaction 1) Performancementioning

confidence: 99%

“…R ADIO Detection And Ranging (Radar) technologies detect objects and movements via electromagnetic or radio waves [1] that can penetrate non-conducting materials, such as glass, wood, plastic, paper, cardboard, and rubber [2], [3]. This capability makes radars appropriate for sensing human gestures in conditions usually not covered by other sensing technologies such as computer vision: environments with poor visibility (e.g., dark places, conditions with fog, rain, dust, smoke, or adverse light) [4], [5], limited accessibility (e.g., hidden or indirect locations, places behind or below rough surfaces) [6], [7], and privacy-sensitivity (e.g., when the user does not want to be filmed or recognized) [8], [9].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Analysis of User-Defined Radar-Based Hand Gestures Sensed Through Multiple Materials

Sluÿters,

Lambot,

Vanderdonckt

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Radar sensing can penetrate non-conducting materials, such as glass, wood, and plastic, which makes it appropriate for recognizing gestures in environments with poor visibility, limited accessibility, and privacy sensitivityWhile the performance of radar-based gesture recognition in these environments has been extensively researched, the preferences that users express for these gestures are less known. To analyze such gestures simultaneously according to their user preference and their system recognition performance, we conducted three gesture elicitation studies each with n 1 =30 participants to identify user-defined, radar-based gestures sensed through three distinct materials: the glass of a shop window, the wood of an office door, and polyvinyl chloride in an emergency. On this basis, we created a new dataset of nine selected gesture classes for n 2 =20 participants repeating twice the same gesture captured by radar through three materials, i.e., glass, wood, and polyvinyl chloride. To uniformly compare recognition rates in these conditions with sensing variations, a specifically tailored procedure was defined and conducted with one-shot radar calibration to train and evaluate a gesture recognizer. 'Wood' achieved the best recognition rate (96.44%), followed by 'Polyvinyl chloride' and 'Glass'. We perform a preference-performance analysis of the gestures by combining the agreement rate from the elicitation studies and the recognition rate from the evaluation.

show abstract

Section: Introductionmentioning

confidence: 71%

Section: ) Agreement Ratementioning

confidence: 99%

Section: A Radar-based Gesture Interaction 1) Performancementioning

confidence: 99%

Section: A Radar-based Gesture Interaction 1) Performancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Analysis of User-Defined Radar-Based Hand Gestures Sensed Through Multiple Materials

Sluÿters,

Lambot,

Vanderdonckt

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

Hand Gesture Recognition for an Off-the-Shelf Radar by Electromagnetic Modeling and Inversion

Sluÿters

Lambot

Vanderdonckt

2022

27th International Conference on Intelligent User Interfaces

View full text Add to dashboard Cite

FORTE: Few Samples for Recognizing Hand Gestures with a Smartphone-attached Radar

Chioccarello,

Sluÿters,

Testolin

et al. 2023

Proc. ACM Hum.-Comput. Interact.

View full text Add to dashboard Cite

Radar sensing technologies offer several advantages over other gesture input modalities, such as the ability to reliably sense human movements, a reasonable deployment cost, insensitivity to ambient conditions such as light, temperature, and the ability to preserve anonymity. These advantages come at the price of high processing complexity mainly due to the spatio-temporal variations of gesture articulation performed by different people. Deep learning methods, such as CNN-LSTM and 3D CNN-LSTM, have a high potential to recognize radar-based gestures but usually require hundreds or thousands of labeled training samples and high processing power. Asking a lot of people to acquire a lot of gestures is particularly tedious and tiring to the point of being unrealistic. To overcome these challenges, we propose FORTE, a hand gesture recognition with few samples based on an optimized CNN architecture working on pre-processed raw data. Using a k=5-fold cross-validation, we define and compare three alternative CNNs for recognizing hand gestures acquired in a semi-mobile context of use with a portable radar attached to a smartphone. The best CNN reaches an accuracy of 94.96% with a precision of 95.92% and a recall of 96.03% for a dataset composed of solely 5 participants producing 2 samples for 20 classes covering 1 pointing, 2 pantomimic, 3 iconic, and 14 semaphoric gestures. We suggest some implications for designing radar-based gestures and we discuss the limitations of this approach.

show abstract

On Indoor Human Sensing Using Commodity Radar

Cited by 4 publications

References 11 publications

Analysis of User-Defined Radar-Based Hand Gestures Sensed Through Multiple Materials

Analysis of User-Defined Radar-Based Hand Gestures Sensed Through Multiple Materials

Hand Gesture Recognition for an Off-the-Shelf Radar by Electromagnetic Modeling and Inversion

FORTE: Few Samples for Recognizing Hand Gestures with a Smartphone-attached Radar

Contact Info

Product

Resources

About