AMNet enables modality-agnostic low-light video enhancement via a Spatial-Spectral Dual-Gated Translator for implicit auxiliary representations and large-scale pretraining on RGB data with synthetic auxiliaries.
What makes for robust multi-modal models in the face of missing modalities? arXiv preprint arXiv:2310.06383, 2023
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
AnyMod-LLVE: Low-Light Video Enhancement with Modality-Agnostic Inference
AMNet enables modality-agnostic low-light video enhancement via a Spatial-Spectral Dual-Gated Translator for implicit auxiliary representations and large-scale pretraining on RGB data with synthetic auxiliaries.