CVJul 21, 2022

Inductive and Transductive Few-Shot Video Classification via Appearance and Temporal Alignments

arXiv:2207.10785v144 citationsh-index: 27Has Code
Originality Incremental advance
AI Analysis

This work addresses the problem of classifying videos with limited labeled data for researchers in computer vision, but it is incremental as it builds on existing few-shot learning approaches.

The authors tackled few-shot video classification by introducing a method that performs appearance and temporal alignments, achieving similar or better results than previous methods on Kinetics and Something-Something V2 datasets.

We present a novel method for few-shot video classification, which performs appearance and temporal alignments. In particular, given a pair of query and support videos, we conduct appearance alignment via frame-level feature matching to achieve the appearance similarity score between the videos, while utilizing temporal order-preserving priors for obtaining the temporal similarity score between the videos. Moreover, we introduce a few-shot video classification framework that leverages the above appearance and temporal similarity scores across multiple steps, namely prototype-based training and testing as well as inductive and transductive prototype refinement. To the best of our knowledge, our work is the first to explore transductive few-shot video classification. Extensive experiments on both Kinetics and Something-Something V2 datasets show that both appearance and temporal alignments are crucial for datasets with temporal order sensitivity such as Something-Something V2. Our approach achieves similar or better results than previous methods on both datasets. Our code is available at https://github.com/VinAIResearch/fsvc-ata.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes