Continual Learning for Food Category Classification Dataset: Enhancing Model Adaptability and Performance

Piyush Kaushik Bhattacharyya; Devansh Tomar; Shubham Mishra; Divyanshu Rai; Yug Pratap Singh; Harsh Yadav; Krutika Verma; Vishal Meena

cs.LGMar 20, 2026

Continual Learning for Food Category Classification Dataset: Enhancing Model Adaptability and Performance

Authors: Piyush Kaushik Bhattacharyya, Devansh Tomar, Shubham Mishra, Divyanshu Rai, Yug Pratap Singh, Harsh Yadav, Krutika Verma, Vishal Meena,

Organizations: School of Computer Engineering, KIIT University, Bhubaneshwar, India

Abstract

Conventional machine learning pipelines often struggle to recognize categories absent from the original trainingset. This gap typically reduces accuracy, as fixed datasets rarely capture the full diversity of a domain. To address this, we propose a continual learning framework for text-guided food classification. Unlike approaches that require retraining from scratch, our method enables incremental updates, allowing new categories to be integrated without degrading prior knowledge. For example, a model trained on Western cuisines could later learn to classify dishes such as dosa or kimchi. Although further refinements are needed, this design shows promise for adaptive food recognition, with applications in dietary monitoring and personalized nutrition planning.

Figures & tables

Parameter	Value
TF-IDF max features	5000
Batch size	32
Epochs	100
Optimizer	Adam
Loss function	Binary Crossentropy
Hidden layer sizes	64, 32

TABLE I: Model Hyperparameters

Model	Accuracy	Loss	Error (MAE)	Precision	Recall	F1 Score	AUC
Logistic Regression	98.63% $\pm$ 0.16%	0.00365 $\pm$ 0.00164	0.00365 $\pm$ 0.00164	98.44% $\pm$ 0.27%	98.83% $\pm$ 0.06%	98.63% $\pm$ 0.16%	98.98% $\pm$ 0.01%
Random Forest	98.89% $\pm$ 0.01%	0.00107 $\pm$ 0.00010	0.00107 $\pm$ 0.00010	98.84% $\pm$ 0.06%	98.03% $\pm$ 0.05%	98.09% $\pm$ 0.01%	98.99% $\pm$ 0.01%
SVM	99.13% $\pm$ 0.01%	0.00064 $\pm$ 0.00034	0.00064 $\pm$ 0.00034	99.88% $\pm$ 0.06%	99.19% $\pm$ 0.01%	99.03% $\pm$ 0.03%	99.12% $\pm$ 0.01%
KNN	96.22% $\pm$ 0.21%	0.03779 $\pm$ 0.00217	0.03779 $\pm$ 0.00217	92.97% $\pm$ 0.37%	100.00% $\pm$ 0.01%	96.35% $\pm$ 0.20%	98.74% $\pm$ 0.09%
Proposed Model	98.38% $\pm$ 2.32%	0.39695 $\pm$ 0.07372	0.01469 $\pm$ 0.02185	96.90% $\pm$ 1.35%	99.36% $\pm$ 0.14%	98.11% $\pm$ 01.35%	98.08% $\pm$ 0.69%

TABLE II: Different classifier’s Performance Comparison with proposed framework

Continual Learning

Continual Learning studies how models acquire knowledge from changing data streams while retaining or selectively revising what they learned before.

See all 304 papers

Dual-Imbalance Continual Learning for Real-World Food Recognition
Mar 31, 2026Xiaoyan Zhang, Jiangpeng HeClass-Incremental Learning Incremental Learning
University of Michigan, Ann Arbor, Michigan, U.S.A · Indiana University, Bloomington, Indiana, U.S.A
Virtual Category-Guided Continual Generalized Category Discovery
Jul 6, 2026Jiahui Xiong, Qiuxia Lai, Hongsong WangGeneralized Category Discovery
School of Computer Science and Engineering, Southeast University, Nanjing 210096, China · State Key Laboratory of Media Convergence and Communication, Communication University of China, Beijing 100024, China · Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, China
Continual Learning without Continual Training
Oct 7, 2026Nikita Narayanan, Ritham Majumdarr, Sonali ParbhooContinual Learning Class-Incremental Learning
Imperial College London

Mar 31, 2026J/K move · Enter open · S save

Dual-Imbalance Continual Learning for Real-World Food Recognition

Xiaoyan Zhang, Jiangpeng He

University of Michigan, Ann Arbor, Michigan, U.S.A · Indiana University, Bloomington, Indiana, U.S.A

Class-Incremental Learning Incremental Learning Image Classification Model Merging

Open paper AskPDF arXiv

Visual food recognition in real-world dietary logging scenarios naturally exhibits severe data imbalance, where a small number of food categories appear frequently while many others occur rarely, resulting in long-tailed class distributions. In practice, food recognition systems often operate in a continual learning setting, where new categories are introduced sequentially over time. However, existing studies typically assume that each incremental step introduces a similar number of new food classes, which rarely happens in real world where the number of newly observed categories can vary significantly across steps, leading to highly uneven learning dynamics. As a result, continual food recognition exhibits a dual imbalance: imbalanced samples within each food class and imbalanced numbers of new food classes to learn at each incremental learning step. In this work, we introduce DIME, a Dual-Imbalance-aware Adapter Merging framework for continual food recognition. DIME learns lightweight adapters for each task using parameter-efficient fine-tuning and progressively integrates them through a class-count guided spectral merging strategy. A rank-wise threshold modulation mechanism further stabilizes the merging process by preserving dominant knowledge while allowing adaptive updates. The resulting model maintains a single merged adapter for inference, enabling efficient deployment without accumulating task-specific modules. Experiments on realistic long-tailed food benchmarks under our step-imbalanced setup show that the proposed method consistently improves by more than 3% over the strongest existing continual learning baselines. Code is available at https://github.com/xiaoyanzhang1/DIME.

Abstract

Continual Learning

Explore similar work

Dual-Imbalance Continual Learning for Real-World Food Recognition

Virtual Category-Guided Continual Generalized Category Discovery

Continual Learning without Continual Training