用LDA压缩冻结的CNN特征，能显著提升粗粒度图像分类准确率。
Supervised Dimensionality Reduction Revisited: Why LDA on Frozen CNN Features Deserves a Second Look
- 在冻结的CNN特征上使用LDA进行有监督降维，保留判别性信息
- 在12组粗粒度任务中11组准确率提升，最高达4.5个百分点，维度降低48%-87%
- 适合粗粒度分类，细粒度任务反而会因丢失细节而表现下降
冻结预训练图像表示广泛用于迁移学习：固定主干网络，提取特征向量，再在顶部训练轻量分类器。该流程通常将完整特征向量输入分类器，即使目标任务类别远少于预训练任务。我们重新审视一种经典替代方案：在线性探测前使用有监督降维（如线性判别分析LDA）。在六种主干网络（ResNet-18、ResNet-50、MobileNetV3-Small、EfficientNet-B0、ViT-B/16、DINOv2-ViT-S/14）上，对CIFAR-100、Tiny ImageNet和CUB-200-2011三个数据集进行评估。在固定逻辑回归协议下，LDA在12个粗粒度配置中11个优于全维特征，准确率最高提升4.5个百分点，特征维度降低48%-87%。但在细粒度的CUB-200上，全特征始终更优。这表明：当类别结构足够粗糙时，均值分离方向可有效捕捉判别信息；但若需精细区分，则过度降维会损失关键线索。我们还对比了PCA、PCA+LDA、正则化LDA、局部Fisher判别分析、邻域成分分析及三种轻量LDA扩展。结果表明，普通LDA在多数粗粒度场景中提供最佳精度-成本权衡，复杂方法极少值得额外开销。本研究为是否在冻结特征分类流程中加入后处理有监督投影提供了明确指导。
原文摘要 · Abstract (English)
Frozen pretrained image representations are widely used for transfer learning: a backbone is kept fixed, feature vectors are extracted, and a lightweight classifier is trained on top. This pipeline usually feeds the full feature vector to the classifier, even when the target task has far fewer classes than the pretraining task. We revisit a classical alternative: supervised dimensionality reduction with Linear Discriminant Analysis (LDA) before linear probing. We evaluate ten dimensionality-reduction strategies on frozen features from six backbones -- ResNet-18, ResNet-50, MobileNetV3-Small, EfficientNet-B0, ViT-B/16, and DINOv2-ViT-S/14 -- across CIFAR-100, Tiny ImageNet, and CUB-200-2011. Under a fixed logistic-regression protocol, LDA improves accuracy over full features in 11 of 12 coarse-grained configurations, with gains up to 4.5 percentage points while reducing feature dimensionality by 48-87%. The same projection consistently hurts on fine-grained CUB-200, where full features win across all six backbones. This establishes a practical boundary condition: LDA is useful when class-level structure is coarse enough to be captured by mean-separating directions, but it can discard subtle cues needed for fine-grained recognition. We also compare LDA with PCA, PCA+LDA, regularized LDA, Local Fisher Discriminant Analysis, Neighbourhood Components Analysis, and three lightweight LDA extensions. The results show that plain LDA offers the best accuracy-cost tradeoff for most coarse-grained settings, while more complex supervised reduction methods rarely justify their additional cost. Overall, the study provides concrete guidance for when post-hoc supervised projection should, and should not, be inserted into frozen-feature image classification pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。