Video Question Answering with Spatio-Temporal Reasoning

68 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

NExT-QA: Next Phase of Question-Answering to Explaining Temporal Actions (2021, methodology): “Regarding visual feature representations, motion feature are important but naively concatenating appearance and motion features usually results in sub-optimal results (refer to EVQA [2], PSAC+ [26] and STVQA [27]).”
Reasoning with Heterogeneous Graph Alignment for Video Question Answering (2020, methodology): “For question, we follow previous work (Gao et al. 2018; Jang et al. 2019; Fan et al. 2019) and represent each word Figure 2: HGA network consists of four stages: representation, fusion, alignment, and reasoning.”
Perception Test: A Diagnostic Benchmark for Multimodal Video Models (2023, background): “Kinetics [36], ActivityNet [10], VGGSound [12], HVU [18], ActivityNet-QA [53], tGIFQA [34]; (ii) Videos collected on demand, filmed by volunteers doing arbitrary activities in indoor or outdoor scenes, e.”
ScanQA: 3D Question Answering for Spatial Scene Understanding (2021, methodology): “[23] proposed a question answering method that takes into account more detailed vision information and motion information by using video.”