CNN-based salient object detection (SOD) methods achieve impressive performance. However, the way semantic information is encoded in them and whether they are category-agnostic is less explored. One major obstacle in studying these questions is the fact that SOD models are built on top of the ImageNet pre-trained backbones which may cause information leakage and feature redundancy. To remedy this, here we first propose an extremely light-weight holistic model tied to the SOD task that can be freed from classification backbones and trained from scratch, and then employ it to study the semantics of SOD models. With the holistic network and representation redundancy reduction by a novel dynamic weight decay scheme, our model has only 100K parameters, ∼ 0.2% of parameters of large models, and performs on par with SOTA on popular SOD benchmarks. Using CSNet, we find that a) SOD and classification methods use different mechanisms, b) SOD models are category insensitive, c) ImageNet pre-training is not necessary for SOD training, and d) SOD models require far fewer parameters than the classification models. The source code is publicly available at https://mmcheng.net/sod100k/.

中文翻译：

---

研究显着对象检测语义的高效模型。

基于 CNN 的显着目标检测 (SOD) 方法取得了令人印象深刻的性能。然而，语义信息在其中编码的方式以及它们是否与类别无关的探索较少。研究这些问题的一个主要障碍是 SOD 模型建立在 ImageNet 预训练主干之上，这可能会导致信息泄漏和特征冗余。为了解决这个问题，我们在这里首先提出了一个与 SOD 任务相关的极轻量级整体模型，该模型可以从分类主干中解放出来并从头开始训练，然后用它来研究 SOD 模型的语义。通过新型动态权重衰减方案的整体网络和表示冗余减少，我们的模型只有 100K 个参数，大约是大型模型参数的 0.2%，并且在流行的 SOD 基准测试中与 SOTA 的性能相当。使用 CSNet，我们发现 a) SOD 和分类方法使用不同的机制，b) SOD 模型对类别不敏感，c) SOD 训练不需要 ImageNet 预训练，d) SOD 模型需要的参数远少于分类模型. 源代码公开于 https://mmcheng.net/sod100k/。