AI Paper Atlas

Research map / Detection and segmentation / CNN features

HBONet: Harmonious Bottleneck on Two Orthogonal Dimensions

Duo Li, Aojun Zhou, Anbang Yao

ICCV 2019 · Detection and segmentation › CNN features · approach: Convolution-transformer backbones · 39 citations (OpenAlex)

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 changedWhereMeasuredEffect
replace MobileNetV2 inverted bottlenecks with HBO modules (width 1.0)MobileNetV2 (1.0) · ImageNet valtop-1 accuracy (%): 72.2 -> 73.1Improved
replace MobileNetV2 bottlenecks with HBO modules at width 0.25MobileNetV2 (0.25) · ImageNet valtop-1 accuracy (%): 52.3 -> 57.3Improved
replace MobileNetV2 bottlenecks with HBO modules at width 0.1MobileNetV2 (0.1) · ImageNet valtop-1 accuracy (%): 34.9 -> 41.5Improved
use HBONet at 224x224 input, width 0.8 vs 0.75MobileNetV2 (0.75) · ImageNet val (224 input)top-1 accuracy (%): 69.8 -> 71.3Improved
use HBONet at 96x96 input (width 0.8)MobileNetV2 (0.75) · ImageNet val (96 input)top-1 accuracy (%): 58.8 -> 61.4Improved
use HBONet at 96x96 input (width 0.35)MobileNetV2 (0.35) · ImageNet val (96 input)top-1 accuracy (%): 45.5 -> 50.3Improved
HBONet 0.6 at 192 input instead of MobileNetV2 0.5 at 224 (similar FLOPs)MobileNetV2 (0.5) · ImageNet valtop-1 accuracy (%): 64.6 -> 67.7Improved
HBONet 0.5 at 224 input instead of MobileNetV2 0.6 at 192 (similar FLOPs)MobileNetV2 (0.6) · ImageNet valtop-1 accuracy (%): 65.6 -> 67.3Improved
use HBONet backbone in SSD320 at width 1.0MobileNetV2 SSD320 · PASCAL VOC 2007 testmAP (%): 70.4 -> 71.0Improved
use HBONet backbone in SSD320 at width 0.1MobileNetV2 SSD320 · PASCAL VOC 2007 testmAP (%): 36.3 -> 42.6Improved
use HBONet backbone for re-ID at width 1.0MobileNetV2 · Market-1501mAP (%): 70.5 -> 74.4Improved
use HBONet backbone for re-ID at width 0.1MobileNetV2 · Market-1501Rank-1 (%): 68.3 -> 73.2Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy