Vision Relation Transformer for Unbiased Scene Graph Generation
This addresses bias and efficiency issues in visual scene understanding for applications like robotics and image captioning, representing a strong incremental advance.
The paper tackles the problem of information loss and bias in Scene Graph Generation (SGG) by introducing VETO with a local-level entity relation encoder and MEET learning strategy, resulting in up to 47% performance improvement and a 10x smaller model size compared to state-of-the-art methods.
Recent years have seen a growing interest in Scene Graph Generation (SGG), a comprehensive visual scene understanding task that aims to predict entity relationships using a relation encoder-decoder pipeline stacked on top of an object encoder-decoder backbone. Unfortunately, current SGG methods suffer from an information loss regarding the entities local-level cues during the relation encoding process. To mitigate this, we introduce the Vision rElation TransfOrmer (VETO), consisting of a novel local-level entity relation encoder. We further observe that many existing SGG methods claim to be unbiased, but are still biased towards either head or tail classes. To overcome this bias, we introduce a Mutually Exclusive ExperT (MEET) learning strategy that captures important relation features without bias towards head or tail classes. Experimental results on the VG and GQA datasets demonstrate that VETO + MEET boosts the predictive performance by up to 47 percentage over the state of the art while being 10 times smaller.