SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning
This work addresses the challenge of classifying unlabelled images from both seen and unseen classes in computer vision, representing an incremental improvement over existing GCD methods.
The paper tackles the problem of Generalized Category Discovery (GCD) by introducing SPTNet, a two-stage adaptation approach that iteratively optimizes model parameters and data parameters, achieving an average accuracy of 61.4% on the SSB benchmark, which surpasses prior state-of-the-art methods by approximately 10%.
Generalized Category Discovery (GCD) aims to classify unlabelled images from both `seen' and `unseen' classes by transferring knowledge from a set of labelled `seen' class images. A key theme in existing GCD approaches is adapting large-scale pre-trained models for the GCD task. An alternate perspective, however, is to adapt the data representation itself for better alignment with the pre-trained model. As such, in this paper, we introduce a two-stage adaptation approach termed SPTNet, which iteratively optimizes model parameters (i.e., model-finetuning) and data parameters (i.e., prompt learning). Furthermore, we propose a novel spatial prompt tuning method (SPT) which considers the spatial property of image data, enabling the method to better focus on object parts, which can transfer between seen and unseen classes. We thoroughly evaluate our SPTNet on standard benchmarks and demonstrate that our method outperforms existing GCD methods. Notably, we find our method achieves an average accuracy of 61.4% on the SSB, surpassing prior state-of-the-art methods by approximately 10%. The improvement is particularly remarkable as our method yields extra parameters amounting to only 0.117% of those in the backbone architecture. Project page: https://visual-ai.github.io/sptnet.