2026-04-20T17:04:27Z - 2026-05-20T17:04:27Z
Overview
24 issues created by 1 user
Opened
#1 Training UI — Merlin-side gesture recorder and test mode (Avocet handles labeling)
Opened
#2 Signal Quality Scorer — SNR + consistency gate for training data
Opened
#3 Learned personal model — local MLP / gradient boost trainer (Phase A+)
Opened
#4 Dwell-to-click — gaze-driven click via configurable hold timer
Opened
#5 System tray mode — background operation after initial configuration
Opened
#6 manage.sh train and test commands
Opened
#7 macOS support — Quartz event tap for OS input injection
Opened
#8 Windows support — win32api SendInput for OS input injection
Opened
#9 Konami code easter egg — training UI brain-wave label mode
Opened
#10 DepthFusion layer — source-agnostic depth adapter interface (cf-core)
Opened
#11 StructuredLight source adapter — OpenNI2 (Xtion/Kinect) and librealsense (RealSense)
Opened
#12 Stereo camera source — OpenCV StereoSGBM + one-time calibration wizard
Opened
#13 Monocular depth source — Depth Anything v2 + MiDaS fallback (PyTorch)
Opened
#14 SteamVR / libsurvive source — Lighthouse 6DoF tracking adapter
Opened
#15 SkeletonFeatures extractor — joint angles and velocities from 3D landmarks
Opened
#16 Linnet event bus integration — share gesture stream with Linnet composition panel
Opened
#17 brainflow integration — CUSTOM_BOARD + OpenBCI Cyton/Ganglion support
Opened
#18 ADS1299 custom board firmware — ESP32 WiFi → brainflow CUSTOM_BOARD protocol
Opened
#19 EEG motor imagery classifier — CSP + LDA (or small CNN) for left/right/feet/rest
Opened
#20 EMG integration — OpenBCI EMG and Myo-compatible armband support
Opened
#21 EEG/EMG signal quality dashboard — per-channel impedance, artifact detection, SNR
Opened
#22 Multi-modal fusion — camera + EEG/EMG combined classification
Opened
#23 BCI adaptive feedback loop — closed-loop neurofeedback for EEG training (Phase C)
Opened
#24 PCB layout — merlin-bci-rev0 (ADS1299 EEG front-end)