arXiv · cs.AI · Oriod Malo (AI Lab UMT), Alexandre Xavier Falc\~ao (IC), Laurent Najman (KUSTAR, LIGM) · 7 小时前模型评分28
Sum12：面向小样本图像分类的轻量多尺度分支编码器
A Novel Lightweight Multi-Scale Branching Encoder for Few-Shot Image Classification
导读
研究者提出轻量卷积编码器 Sum12，用七个并行卷积分支（卷积核尺寸之和为 12）改进原型网络（Prototypical Networks）的嵌入阶段，在 Omniglot、CIFAR-FS、FC100 和 miniImageNet 的小样本协议下普遍优于原始 ProtoNet 编码器，同时保持紧凑模型体积。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
应用领域：医疗与生命科学
事件类型：研究发布
证据摘录
- “This paper presents the Sum12 Multi-Scale branching encoder, a lightweight convolutional architecture designed to improve the embedding stage of Prototypical Networks without substantially increasing computational cost.”
- “Preprint version.”
- “However, complete end-to- end validation on the target embedded platform, including resolution of the observed implementation discrepancies, remains future work.”
来源：arXiv · cs.AI · arxiv.org