STCM-Net: A symmetrical one-stage network for temporal language localization in videos

The Elements of Temporal Sentence Grounding in Videos: A Survey and Future Directions

“…In this sense, multimodal interaction is overlooked. To remedy, PLN [89], SMIN [53], CLEAR [90], and STCM-Net [91] disentangle video proposals into different temporal granularities [89], [91] or different semantic contents [53], [90], and perform crossmodal reasoning at both coarse-and fine-grained granularities. VLG-Net [92] and RaNet [54] maintain query words and video proposals in a graph, and adopt GCN [4], [93] to conduct intra-and inter-modal interactions for cross-modal reasoning.…”

Towards temporal sentence grounding in videos

“…In this sense, multimodal interaction is overlooked. To remedy, PLN [79], SMIN [57], CLEAR [80], and STCM-Net [81] disentangle video proposals into different temporal granularities [79,81] or different semantic contents [57,80], and perform cross-modal reasoning at both coarse-and fine-grained granularities. VLG-Net [82] and RaNet [58] maintain query words and video proposals in a graph, and adopt GCN [83,84] to conduct intra-and inter-modal interactions for cross-modal reasoning.…”