X/Y/Z coordinates per frame
Three-dimensional position data at full temporal resolution from triangulated multi-camera reconstruction. Calibration residuals reported for quality assessment.
Three-dimensional behavioral trajectories from multi-camera recordings
Reconstruct vertical movement — rearing height, climbing behavior, jump trajectories — from calibrated camera arrays.
Standard top-down or side-view tracking projects three-dimensional behavior into a flat plane. Rearing is classified as a binary event (present/absent) rather than measured by height. Climbing behavior is unquantified. Jumping is detected only as a position discontinuity, losing trajectory information.
ConductVision triangulates position from multiple synchronized camera views after a straightforward calibration procedure. X/Y/Z coordinates are computed at every frame, enabling quantitative measures of rearing height, vertical displacement, and three-dimensional path analysis.
Mouse recording with paired camera views and reconstructed 3D keypoints in the same frame. ConductVision combines calibrated camera geometry, frame-synchronized keypoints, and trajectory refinement to produce spatial tracking evidence.
Three-dimensional position data at full temporal resolution from triangulated multi-camera reconstruction. Calibration residuals reported for quality assessment.
Total 3D path length, cumulative vertical displacement, and rearing height distribution. Decomposed into horizontal and vertical components.
Rendered 3D trajectory overlaid on synchronized camera views. Rotatable visualization for presentations and publications.
Measure rearing as a continuous variable — height, duration, and frequency — rather than a binary event. Distinguish supported (wall) from unsupported (free) rearing by spatial context.
Repetitive vertical movements — head bobbing, vertical circling — are captured in full 3D trajectories. Pattern analysis identifies stereotypic sequences invisible in 2D.
Quantify climbing speed, height reached, and time at elevation in wire mesh or enriched cage environments. Detect progressive motor decline in longitudinal studies.
Track navigation through multi-level or complex 3D environments. Compute volumetric exploration, level transitions, and spatial memory indices.
| Feature | ConductVision | Typical systems |
|---|---|---|
| Dimensions tracked | X, Y, Z — full 3D | X, Y only |
| Vertical resolution | Continuous height measurement | Binary rearing detection |
| Calibration | Guided multi-camera calibration | Single camera, no calibration |
| 3D visualization | Rotatable 3D trajectory render | Not available |
| Vertical behavior types | Rearing, climbing, jumping | Rearing only (binary) |
High-resolution 30 fps tracking that captures sub-second behavioral events conventional systems miss.

Multi-keypoint body part tracking — head, paws, tail, body center — with pretrained and custom model support.
Persistent markerless identity tracking for up to four animals simultaneously without physical markers or implants.
See how 3D reconstruction reveals behavioral events that 2D tracking cannot detect.