Name
Abstract | ||
|---|---|---|
Feature reassembly, i.e. feature downsampling and upsampling, is a key operation in a number of modern convolutional network architectures, e.g., residual networks and feature pyramids. Its design is critical for dense prediction tasks such as object detection and semantic/instance segmentation. In this work, we propose unified Content-Aware ReAssembly of FEatures (CARAFE++), a universal, lightweight, and highly effective operator to fulfill this goal. CARAFE++ has several appealing properties: (1) Unlike conventional methods such as pooling and interpolation that only exploit sub-pixel neighborhood, CARAFE++ aggregates contextual information within a large receptive field. (2) Instead of using a fixed kernel for all samples (e.g. convolution and deconvolution), CARAFE++ generates adaptive kernels on-the-fly to enable instance-specific content-aware handling. (3) CARAFE++ introduces little computational overhead and can be readily integrated into modern network architectures. We conduct comprehensive evaluations on standard benchmarks in object detection, instance/semantic segmentation, and image inpainting. CARAFE++ shows consistent and substantial gains on mainstream methods across all the tasks with negligible computational overhead. It shows great potential to serve as a strong building block for modern deep networks. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/TPAMI.2021.3074370 | IEEE Transactions on Pattern Analysis and Machine Intelligence |
Keywords | DocType | Volume |
Feature reassembly,object detection,instance segmentation,semantic segmentation,image inpainting | Journal | 44 |
Issue | ISSN | Citations |
9 | 0162-8828 | 0 |
PageRank | References | Authors |
0.34 | 24 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jiaqi Wang | 1 | 77 | 4.20 |
| Kai Chen | 2 | 130 | 8.65 |
| Rui Xu | 3 | 20 | 2.66 |
| Ziwei Liu | 4 | 1361 | 63.23 |
| Chen Change Loy | 5 | 4484 | 178.56 |
| Dahua Lin | 6 | 1117 | 72.62 |