Event Camera (DVS) — intensity reconstruction from asynchronous events L1-078
Unclaimed Principle — open for contribution
This Principle is declared in the catalog but has no reference solver, no pinned dataset, and is not registered on-chain. There is no reward pool. Submitting a cert against this Principle today will record the cert for reproducibility but pay zero PWM.
To claim it as a Bounty #7 contribution: open a PR adding (1) a reference solver, (2) ≥1 dataset pinned to IPFS, (3) updates to the L3 manifest with dataset CIDs. After verifier-agent triple-review, the founders' 3-of-5 multisig signs PWMRegistry.register() and the Principle becomes mineable.
Forward model E
Each pixel of an event sensor (DVS / DAVIS) independently emits an event e = (t, u, v, p) with polarity p = +/-1 whenever the log-intensity change since the last event at that pixel crosses a contrast threshold C. The asynchronous event stream encodes temporal brightness changes at microsecond resolution but not absolute intensity. Reconstructing a standard intensity video I(u,v,t) from the event stream is an ill-posed temporal inverse problem.
L-DAG
Well-posedness W
- Existence:
- true
- Uniqueness:
- false
- Stability:
- conditional
- κ:
- 4000
Intensity reconstruction is ill-posed (DC level is unobservable; only log-differentials are sampled). Unique reconstruction requires an anchor frame (DAVIS hybrid output) or a strong prior (video smoothness, learned models). Contrast-threshold variance sigma_C and noise events add strong instability at low event rates.
Solvability C
- Solver class:
- E2VID (recurrent neural net), FireNet (U-Net), EV-Gait (Poisson-Events), event-to-frame reconstruction with optical-flow prior
- Convergence rate q:
- 1.5
- Complexity:
- O(N_events) for async integration; O(H * W * N_t) for recurrent reconstruction