arrow_backBack to questions View categoryarrow_forward

Sensing, Recognition, and Prediction Methods / Activity Recognition, Segmentation, and Identity Discrimination Methods

Users' idle attention moments are difficult to identify, missing opportunities to improve experience.

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

UbiComp '21Killing-Time Detection from Smartphone Screenshots

Similar questions

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

Indoor action recognition lacks large-scale, multi-category open datasets.

UbiComp '24XRF55: A Radio Frequency Dataset for Human Indoor Action Analysis

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

WiFi sensing is limited by data scarcity, restricting application scale.

UbiComp '24RFBoost: Understanding and Boosting Deep WiFi Sensing via Physical Data Augmentation

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

Current wearables struggle to accurately recognize complex activities such as running or drinking water.

UbiComp '24HyperHAR: Inter-sensing Device Bilateral Correlations and Hyper-correlations Learning Approach for Wearable Sensing Device Based Human Activity Recognition

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

Technology design overemphasizes goal completion and neglects the deep meaning of human activity itself.

CHI '24Searching for the Non-Consequential: Dialectical Activities in HCI and the Limits of Computers

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

Existing WiFi recognition systems are impractical due to environmental changes and struggle to identify new users.

UbiComp '22MetaGanFi: Cross-Domain Unseen Individual Identification Using WiFi Signals

lightbulbPractical problemActivity Recognition, Segmentation, and Identity Discrimination Methods

In team or multi-user scenarios, action recognition systems struggle to precisely distinguish individual user behavior.

UbiComp '22IMar: Multi-user Continuous Action Recognition with WiFi Signals

Users' idle attention moments are difficult to identify, missing opportunities to improve experience.

Similar questions

Related papers

XRF55: A Radio Frequency Dataset for Human Indoor Action Analysis

RFBoost: Understanding and Boosting Deep WiFi Sensing via Physical Data Augmentation

Searching for the Non-Consequential: Dialectical Activities in HCI and the Limits of Computers