AI Paper Atlas

Research map / Video / Video segmentation and tracking

Diagnosing and Preventing Instabilities in Recurrent Video Processing

Thomas Tanay, Aivar Sootla, Matteo Maggioni, Puneet K. Dokania, Philip Torr, Ales Leonardis, Gregory Slabaugh

TPAMI 2022 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 10 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
add Hard Lipschitz constraint SRN-C-1.0-1.0 (spectral norm 1) (ablation)VResNet-feat · Vimeo-90k denoisingPSNR_7 (dB): 35.74 -> 35.31Hurt
add Soft Lipschitz constraint SRN-C-2.0-0.1 (ablation)VResNet-feat · Vimeo-90k denoisingPSNR_7 (dB): 35.74 -> 35.59 (-0.15 dB)Hurt
use Soft instead of Hard Lipschitz constraint (alpha=2.0, beta=0.1 vs alpha=1.0, beta=1.0) (ablation)SRN-C on VResNet-feat · Vimeo-90k denoisingPSNR_7 (dB): 35.31 -> 35.59 (+0.18dB)Improved
lower alpha from 2.0 to 1.0 in SRN-C (beta=1.0) (ablation)VResNet-feat · Vimeo-90k denoisingPSNR_7 (dB): 35.71 -> 35.31 (-0.4dB)Hurt
use SRN-C instead of SRN (normalise the conv layer not the reshaped kernel) (ablation)VResNet-feat · Vimeo-90k denoising stabilityspectral norm of conv layers: around 2.5 (±0.2) for SRN -> set precisely to alphaImproved
reduce alpha below 1 (SRN-C, beta=1.0) (ablation)VResNet-feat · Vimeo-90k denoisinglong-sequence stability: alpha>1 unstable, alpha<1 stableImproved
reduce stable rank parameter beta below 0.1 (alpha=2.0) (ablation)VResNet-feat · Vimeo-90k denoisinglong-sequence stability: beta>0.1 unstable, beta<0.1 stableImproved
add SRN-C-2.0-0.05 soft constraintVResNet-feat · Vimeo-90k 2x super-resolutionPSNR_7 (dB): 32.58 -> 32.64 (+0.06 dB)Improved
add SRN-C-1.0-1.0 hard constraintVResNet-feat · Vimeo-90k 2x super-resolutionPSNR_7 (dB): 32.58 -> 32.57No clear effect
add feature dampening lambda=0.55 at inference (ablation)VResNet-feat · Vimeo-90k denoisingPSNR_7 (dB): 35.74 -> 34.62 (-1.12 dB)Hurt
add recurrent connection (RLSP) to feedforward VResNet multi-frame models (ablation)VResNet · Vimeo-90k denoisingPSNR: about 0.5dBImproved
make RLSP recurrence bidirectional (ablation)VResNet-RLSP · Vimeo-90k denoisingPSNR: another 0.4dBImproved

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