AI Paper Atlas

Research map / Video / Video segmentation and tracking

SoDaCam: Software-defined Cameras via Single-Photon Imaging

Varun Sundar, Andrei Ardelean, Tristan Swedish, Claudio Bruschini, Edoardo Charbon, Mohit Gupta

ICCV 2023 · Video › Video segmentation and tracking · approach: Spike and neuromorphic deblurring · 20 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
compute projections near-sensor on UltraPhase instead of reading out the full photon-cubeSoDaCam (SwissSPAD2) · UltraPhase 24x12 array, 40 Hz readoutreadout bandwidth (kbps): 28125 -> 135Improved
compute projections near-sensor instead of full photon-cube readoutSoDaCam (SwissSPAD2) · UltraPhase 24x12 array, 40 Hz readouttotal power (uW): 1518.8 -> 28.6 (three projections)Improved
emulate cameras under ambient light compared with CMOS at 40 FPSSoDaCam (SwissSPAD2) · 512x256 array, ambient lighttotal power (mW): 66.8 vs ~15-30 (CMOS)Hurt
code only dynamic regions in multi-bucket captures (ablation)multi-bucket compressive capture · video compressive sensingbandwidth relative to single snapshot: 1.45x bandwidthImproved
use multi-bucket captures instead of single compressive snapshot (ablation)high-speed compressive imaging · 80-frame reconstruction from 25 Hz snapshotsvideo reconstruction qualityImproved
use SPAD response curve instead of log-MLE brightness encoding for events (ablation)SPAD event emulation · low-light event imagingevent qualityMixed
use conventional high-speed camera (Infinicam, 4 kHz) instead of SPAD for photon-cube projectionsSoDaCam projections · ambient light comparisonprojection qualityHurt
emulate events with SPAD versus Prophesee EVK4 under low lightSPAD-events · low-light event imagingevent image qualityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy