Research map / Architectures and efficiency / Attention and transformers
ParCNetV2: Oversized Kernel with Enhanced Attention
Ruihan Xu, Haokui Zhang, Wenze Hu, Shiliang Zhang, Xiaoyu Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add oversized convolution (ablation) | ParCNetV2-XT · ImageNet-1K val | top-1 accuracy (%): 78.9 -> 79.4 | Improved |
| add spatial bifurcate gate unit (ablation) | ParCNetV2-XT · ImageNet-1K val | top-1 accuracy (%): 79.2 -> 79.4 | Improved |
| add channel bifurcate gate unit (ablation) | ParCNetV2-XT · ImageNet-1K val | top-1 accuracy (%): 79.1 -> 79.4 | Improved |
| use uniform local-global convolution block in all stages (ablation) | ParCNetV2-XT · ImageNet-1K val | top-1 accuracy (%): 79.2 -> 79.4 | Improved |
| replace Swin-T/ConvNeXt-T backbone with ParCNetV2-T | Mask R-CNN 3x · COCO | AP bbox: 46.2 -> 48.9 | Improved |
| replace ConvNeXt-T backbone with ParCNetV2-T | UperNet · ADE20K (multi-scale) | mIoU (%): 46.7 -> 49.4 | Improved |
| replace Swin-T with ParCNetV2-T | RK3288 inference · ImageNet-1K / RK3288 | latency (ms): 855 -> 840 | Improved |
Closest papers by meaning
- ParC-Net: Position Aware Circular Convolution with Merits from ConvNets and Transformer (ECCV 2022 · similarity 0.89)
- CAT: Circular-Convolutional Attention for Sub-Quadratic Transformers (NeurIPS 2025 · similarity 0.88)
- More ConvNets in the 2020s: Scaling up Kernels Beyond 51x51 using Sparsity (ICLR 2023 · similarity 0.87)
- LSNet: See Large, Focus Small (CVPR 2025 · similarity 0.86)
- LookHere: Vision Transformers with Directed Attention Generalize and Extrapolate (NeurIPS 2024 · similarity 0.86)
- AsCAN: Asymmetric Convolution-Attention Networks for Efficient Recognition and Generation (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
