paper-with-me

Papers

Implicit Temporal Modeling with Learnable Alignment for Video Recognition

2023-04-20 · ICCV 2023 1 · Shuyuan Tu, Qi Dai, Zuxuan Wu, Zhi-Qi Cheng, Han Hu, Yu-Gang Jiang

Contrastive language-image pretraining (CLIP) has demonstrated remarkable success in various image tasks. However, how to extend CLIP with effective temporal modeling is still an open and crucial problem. Existing factorized or joint spatial-temporal modeling trades off between the efficiency and performance. While modeling temporal information within straight through tube is widely adopted in literature, we find that simple frame alignment already provides enough essence without temporal attention. To this end, in this paper, we proposed a novel Implicit Learnable Alignment (ILA) method, which minimizes the temporal modeling effort while achieving incredibly high performance. Specifically, for a frame pair, an interactive point is predicted in each frame, serving as a mutual information rich region. By enhancing the features around the interactive point, two frames are implicitly aligned. The aligned features are then pooled into a single token, which is leveraged in the subsequent spatial self-attention. Our method allows eliminating the costly or insufficient temporal self-attention in video. Extensive experiments on benchmarks demonstrate the superiority and generality of our module. Particularly, the proposed ILA achieves a top-1 accuracy of 88.7% on Kinetics-400 with much fewer FLOPs compared with Swin-L and ViViT-H. Code is released at https://github.com/Francis-Rings/ILA .

📄 PDF Abstract BibTeX arXiv:2304.10465

Code (1)

francis-rings/ila 공식 구현 pytorch

Tasks

Action ClassificationAction RecognitionVideo Recognition

Methods 이 논문이 사용한 방법론

CLIP Contrastive Language-Image Pre-training (CLIP), consisting of a simplified version of ConVIRT trained from scratch, is an efficient method of image representation learning…

Similar Papers 제목 키워드 기반

Learning Implicit Temporal Alignment for Few-shot Video Classification

2021-05-11 · Songyang Zhang, Jiale Zhou, Xuming He

Few-shot video classification aims to learn new video categories with only a few labeled examples, alleviating the burden of costly annotation in real-world applications. However, it is particularly challenging to learn …

Action Recognition In VideosClassificationFew-Shot LearningVideo Classification

MLVTG: Mamba-Based Feature Alignment and LLM-Driven Purification for Multi-Modal Video Temporal Grounding

2025-06-10 · Zhiyi Zhu, Xiaoyu Wu, Zihao Liu, Linlin Yang

Video Temporal Grounding (VTG), which aims to localize video clips corresponding to natural language queries, is a fundamental yet challenging task in video understanding. Existing Transformer-based methods often suffer …

Language ModelingLanguage ModellingLarge Language ModelMamba+2

LAMV: Learning to Align and Match Videos With Kernelized Temporal Layers

2018-06-01 · CVPR 2018 6 · Lorenzo Baraldi, Matthijs Douze, Rita Cucchiara, Hervé Jégou

This paper considers a learnable approach for comparing and aligning videos. Our architecture builds upon and revisits temporal match kernels within neural networks: we propose a new temporal layer that finds temporal al…

Copy DetectionRetrievalTripletVideo Alignment+1

When and What: Diffusion-Grounded VideoLLM with Entity Aware Segmentation for Long Video Understanding

2025-08-21 · Pengcheng Fang, Yuxia Chen, Rui Guo arxiv

Understanding videos requires more than answering open ended questions, it demands the ability to pinpoint when events occur and how entities interact across time. While recent Video LLMs have achieved remarkable progres…

VTaMo: Video-Text Alignment Model for Sign Language Translation

2026-07-10 · Junyi Hu, Zhewen He, Haomian Huang, Aoxiang Yang 외 arxiv

Sign language translation (SLT) converts continuous sign videos into spoken language text. Gloss-free approaches leverage pre-trained visual encoders and language models but rely on implicit cross-modal alignment from tr…

Sign Language TranslationContrastive Learning