# Weakly Supervised Video Scene Graph Generation via Natural Language Supervision arXiv: https://arxiv.org/abs/2502.15370 ยท HTML: https://mlsanity.com/paper/2502.15370/ ## Key facts - Cited by 4 papers in the map (data through 14 Oct 2026). - Builds on 42 of its 60 references that are in the map. - First posted on arXiv 21 Feb 2025. ## Builds on (most cited first, 30 of 42) - [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021) - [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020) - [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022) - [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023) - [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016) - [Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression](https://mlsanity.com/paper/1902.09630/) (2019) - [VideoChat: Chat-Centric Video Understanding](https://mlsanity.com/paper/2305.06355/) (2023) - [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017) - [CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval](https://mlsanity.com/paper/2104.08860/) (2021) - [InternVideo: General Video Foundation Models via Generative and Discriminative Learning](https://mlsanity.com/paper/2212.03191/) (2022) - [Auto-Encoding Scene Graphs for Image Captioning](https://mlsanity.com/paper/1812.02378/) (2018) - [Action Genome: Actions as Composition of Spatio-temporal Scene Graphs](https://mlsanity.com/paper/1912.06992/) (2019) - [Controlling Linguistic Style Aspects in Neural Language Generation](https://mlsanity.com/paper/1707.02633/) (2017) - [Spatial-Temporal Transformer for Dynamic Scene Graph Generation](https://mlsanity.com/paper/2107.12309/) (2021) - [Dense and Aligned Captions (DAC) Promote Compositional Reasoning in VL Models](https://mlsanity.com/paper/2305.19595/) (2023) - [UnLoc: A Unified Framework for Video Localization Tasks](https://mlsanity.com/paper/2308.11062/) (2023) - [Learning to Generate Scene Graph from Natural Language Supervision](https://mlsanity.com/paper/2109.02227/) (2021) - [RLIP: Relational Language-Image Pre-training for Human-Object Interaction Detection](https://mlsanity.com/paper/2209.01814/) (2022) - [Target Adaptive Context Aggregation for Video Scene Graph Generation](https://mlsanity.com/paper/2108.08121/) (2021) - [RLIPv2: Fast Scaling of Relational Language-Image Pre-training](https://mlsanity.com/paper/2308.09351/) (2023) - [Video Relationship Reasoning using Gated Spatio-Temporal Energy Graph](https://mlsanity.com/paper/1903.10547/) (2019) - [Unbiased Scene Graph Generation in Videos](https://mlsanity.com/paper/2304.00733/) (2023) - [Structured Query-Based Image Retrieval Using Scene Graphs](https://mlsanity.com/paper/2005.06653/) (2020) - [Weakly Supervised Temporal Action Localization via Representative Snippet Knowledge Propagation](https://mlsanity.com/paper/2203.02925/) (2022) - [Exploiting Long-Term Dependencies for Generating Dynamic Scene Graphs](https://mlsanity.com/paper/2112.09828/) (2021) - [LLM4SGG: Large Language Models for Weakly Supervised Scene Graph Generation](https://mlsanity.com/paper/2310.10404/) (2023) - [Linguistic Structures as Weak Supervision for Visual Scene Graph Generation](https://mlsanity.com/paper/2105.13994/) (2021) - [Weakly Supervised Visual Semantic Parsing](https://mlsanity.com/paper/2001.02359/) (2020) - [Zero-shot Visual Relation Detection via Composite Visual Cues from Large Language Models](https://mlsanity.com/paper/2305.12476/) (2023) - [Generating Natural Language Explanations for Visual Question Answering using Scene Graphs and Visual Attention](https://mlsanity.com/paper/1902.05715/) (2019) ## Built on by (most cited first, 4 of 4) - [SAMJAM: Zero-Shot Video Scene Graph Generation for Egocentric Kitchen Videos](https://mlsanity.com/paper/2504.07867/) (2025) - [TRKT: Weakly Supervised Dynamic Scene Graph Generation with Temporal-enhanced Relation-aware Knowledge Transferring](https://mlsanity.com/paper/2508.04943/) (2025) - [Revisiting Weakly-Supervised Video Scene Graph Generation via Pair Affinity Learning](https://mlsanity.com/paper/2603.21559/) (2026) - [SAGE-OR: Semi-supervised Adaptive Scene Graph Generation for Operating Rooms](https://mlsanity.com/paper/2608.15336/) (2026) ## Abstract (arXiv) Existing Video Scene Graph Generation (VidSGG) studies are trained in a fully supervised manner, which requires all frames in a video to be annotated, thereby incurring high annotation cost compared to Image Scene Graph Generation (ImgSGG). Although the annotation cost of VidSGG can be alleviated by adopting a weakly supervised approach commonly used for ImgSGG (WS-ImgSGG) that uses image captions, there are two key reasons that hinder such a naive adoption: 1) Temporality within video captions, i.e., unlike image captions, video captions include temporal markers (e.g., before, while, then, after) that indicate time related details, and 2) Variability in action duration, i.e., unlike human actions in image captions, human actions in video captions unfold over varying duration. To address these issues, we propose a Natural Language-based Video Scene Graph Generation (NL-VSGG) framework that only utilizes the readily available video captions for training a VidSGG model. NL-VSGG consists of two key modules: Temporality-aware Caption Segmentation (TCS) module and Action Duration Variability-aware caption-frame alignment (ADV) module. Specifically, TCS segments the video captions into multiple sentences in a temporal order based on a Large Language Model (LLM), and ADV aligns each segmented sentence with appropriate frames considering the variability in action duration. Our approach leads to a significant enhancement in performance compared to simply applying the WS-ImgSGG pipeline to VidSGG on the Action Genome dataset. As a further benefit of utilizing the video captions as weak supervision, we show that the VidSGG model trained by NL-VSGG is able to predict a broader range of action classes that are not included in the training data, which makes our framework practical in reality.