Effective Scene Graph Generation by Statistical Relation Distillation

Thanh-Son Nguyen, Hong Yang, Basura Fernando; Proceedings of the Winter Conference on Applications of Computer Vision (WACV), 2025, pp. 8409-8419

Abstract


Annotating scene graphs for images is a time-consuming task resulting in many instances of missing relations within existing datasets. In this paper we introduce the Statistical Relation Distillation (SRD) method to enhance scenegraph datasets. SRD leverages human-annotated relations alongside object-to-object and predicate-to-predicate similarities to reinforce the existence likelihood of scene graph relations. Moreover SRD can augment relational frequency using relations of non-selected object and predicate categories that are usually omitted by scene graph generation (SGG) tasks. The output from SRD derives the prior probability which is combined with model-predicted probabilities to annotate missing relations in training images and sub-sequently re-train SGG models on the augmented dataset. We evaluate our proposed method on Visual Genome and GQA-200 datasets. Experimental results show that training on the augmented dataset enhances the performance of prominent scene-graph generation models. The implementation code is at https://github.com/LUNAProject22/SRD.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Nguyen_2025_WACV, author = {Nguyen, Thanh-Son and Yang, Hong and Fernando, Basura}, title = {Effective Scene Graph Generation by Statistical Relation Distillation}, booktitle = {Proceedings of the Winter Conference on Applications of Computer Vision (WACV)}, month = {February}, year = {2025}, pages = {8409-8419} }