MoSA improves dynamic scene graph generation by fusing motion attributes with spatial features and aligning them cross-modally with relationship text embeddings, plus a weighted loss for rare classes, achieving top results on Action Genome.
Dynamic scene graph generation via anticipatory pre-training,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
FReMuRe improves recall of long-tail relationships in video scene graphs via relation-specific branches, frequency-aware predicate embeddings, and new Bayesian/GMM classification heads, with reported gains on the Action Genome dataset.
citing papers explorer
-
MOSA: Motion-Guided Semantic Alignment for Dynamic Scene Graph Generation
MoSA improves dynamic scene graph generation by fusing motion attributes with spatial features and aligning them cross-modally with relationship text embeddings, plus a weighted loss for rare classes, achieving top results on Action Genome.
-
Frequency-guided Multi-level Reasoning for Scene Graph Generation in Video
FReMuRe improves recall of long-tail relationships in video scene graphs via relation-specific branches, frequency-aware predicate embeddings, and new Bayesian/GMM classification heads, with reported gains on the Action Genome dataset.