Foundation Model-Powered 3D Few-Shot Class Incremental Learning via Training-Free Adaptor

Sahar Ahmadi, Ali Cheraghian*, Morteza Saberi, Md Towsif Abir, Hamidreza Dastmalchi, Farookh Hussain, Shafin Rahman

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Recent advances in deep learning for processing point clouds hold increased interest in Few-Shot Class Incremental Learning (FSCIL) for 3D computer vision. This paper introduces a new method to tackle the Few-Shot Continual Incremental Learning (FSCIL) problem in 3D point cloud environments. We leverage a foundational 3D model trained extensively on point cloud data. Drawing from recent improvements in foundation models, known for their ability to work well across different tasks, we propose a novel strategy that does not require additional training to adapt to new tasks. Our approach uses a dual cache system: first, it uses previous test samples based on how confident the model was in its predictions to prevent forgetting, and second, it includes a small number of new task samples to prevent overfitting. This dynamic adaptation ensures strong performance across different learning tasks without needing lots of fine-tuning. We tested our approach on datasets like ModelNet, ShapeNet, ScanObjectNN, and CO3D, showing that it outperforms other FSCIL methods and demonstrating its effectiveness and versatility. The code is available at https://github.com/ahmadisahar/ACCV_FCIL3D.

Original languageEnglish
Title of host publicationComputer Vision – ACCV 2024 - 17th Asian Conference on Computer Vision, Proceedings
EditorsMinsu Cho, Ivan Laptev, Du Tran, Angela Yao, Hongbin Zha
PublisherSpringer Science and Business Media Deutschland GmbH
Pages178-195
Number of pages18
ISBN (Print)9789819609710
DOIs
Publication statusPublished - 2025
Event17th Asian Conference on Computer Vision, ACCV 2024 - Hanoi, Viet Nam
Duration: 8 Dec 202412 Dec 2024

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume15481 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference17th Asian Conference on Computer Vision, ACCV 2024
Country/TerritoryViet Nam
CityHanoi
Period8/12/2412/12/24

Fingerprint

Dive into the research topics of 'Foundation Model-Powered 3D Few-Shot Class Incremental Learning via Training-Free Adaptor'. Together they form a unique fingerprint.

Cite this