Picture for Masakazu Yoshimura

Masakazu Yoshimura

Mixed-precision Supernet Training from Vision Foundation Models using Low Rank Adapter

Mar 29, 2024
Figure 1 for Mixed-precision Supernet Training from Vision Foundation Models using Low Rank Adapter
Figure 2 for Mixed-precision Supernet Training from Vision Foundation Models using Low Rank Adapter
Figure 3 for Mixed-precision Supernet Training from Vision Foundation Models using Low Rank Adapter
Figure 4 for Mixed-precision Supernet Training from Vision Foundation Models using Low Rank Adapter
Viaarxiv icon

PQDynamicISP: Dynamically Controlled Image Signal Processor for Any Image Sensors Pursuing Perceptual Quality

Mar 15, 2024
Figure 1 for PQDynamicISP: Dynamically Controlled Image Signal Processor for Any Image Sensors Pursuing Perceptual Quality
Figure 2 for PQDynamicISP: Dynamically Controlled Image Signal Processor for Any Image Sensors Pursuing Perceptual Quality
Figure 3 for PQDynamicISP: Dynamically Controlled Image Signal Processor for Any Image Sensors Pursuing Perceptual Quality
Figure 4 for PQDynamicISP: Dynamically Controlled Image Signal Processor for Any Image Sensors Pursuing Perceptual Quality
Viaarxiv icon

Self-Supervised Reversed Image Signal Processing via Reference-Guided Dynamic Parameter Selection

Mar 24, 2023
Figure 1 for Self-Supervised Reversed Image Signal Processing via Reference-Guided Dynamic Parameter Selection
Figure 2 for Self-Supervised Reversed Image Signal Processing via Reference-Guided Dynamic Parameter Selection
Figure 3 for Self-Supervised Reversed Image Signal Processing via Reference-Guided Dynamic Parameter Selection
Figure 4 for Self-Supervised Reversed Image Signal Processing via Reference-Guided Dynamic Parameter Selection
Viaarxiv icon

Efficient Joint Detection and Multiple Object Tracking with Spatially Aware Transformer

Nov 09, 2022
Figure 1 for Efficient Joint Detection and Multiple Object Tracking with Spatially Aware Transformer
Figure 2 for Efficient Joint Detection and Multiple Object Tracking with Spatially Aware Transformer
Figure 3 for Efficient Joint Detection and Multiple Object Tracking with Spatially Aware Transformer
Figure 4 for Efficient Joint Detection and Multiple Object Tracking with Spatially Aware Transformer
Viaarxiv icon

DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition

Nov 02, 2022
Figure 1 for DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition
Figure 2 for DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition
Figure 3 for DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition
Figure 4 for DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition
Viaarxiv icon

Rawgment: Noise-Accounted RAW Augmentation Enables Recognition in a Wide Variety of Environments

Oct 28, 2022
Figure 1 for Rawgment: Noise-Accounted RAW Augmentation Enables Recognition in a Wide Variety of Environments
Figure 2 for Rawgment: Noise-Accounted RAW Augmentation Enables Recognition in a Wide Variety of Environments
Figure 3 for Rawgment: Noise-Accounted RAW Augmentation Enables Recognition in a Wide Variety of Environments
Figure 4 for Rawgment: Noise-Accounted RAW Augmentation Enables Recognition in a Wide Variety of Environments
Viaarxiv icon

MBAPose: Mask and Bounding-Box Aware Pose Estimation of Surgical Instruments with Photorealistic Domain Randomization

Mar 15, 2021
Figure 1 for MBAPose: Mask and Bounding-Box Aware Pose Estimation of Surgical Instruments with Photorealistic Domain Randomization
Figure 2 for MBAPose: Mask and Bounding-Box Aware Pose Estimation of Surgical Instruments with Photorealistic Domain Randomization
Figure 3 for MBAPose: Mask and Bounding-Box Aware Pose Estimation of Surgical Instruments with Photorealistic Domain Randomization
Figure 4 for MBAPose: Mask and Bounding-Box Aware Pose Estimation of Surgical Instruments with Photorealistic Domain Randomization
Viaarxiv icon

FOSS: Multi-Person Age Estimation with Focusing on Objects and Still Seeing Surroundings

Oct 15, 2020
Figure 1 for FOSS: Multi-Person Age Estimation with Focusing on Objects and Still Seeing Surroundings
Figure 2 for FOSS: Multi-Person Age Estimation with Focusing on Objects and Still Seeing Surroundings
Figure 3 for FOSS: Multi-Person Age Estimation with Focusing on Objects and Still Seeing Surroundings
Figure 4 for FOSS: Multi-Person Age Estimation with Focusing on Objects and Still Seeing Surroundings
Viaarxiv icon

Single-Shot Pose Estimation of Surgical Robot Instruments' Shafts from Monocular Endoscopic Images

Mar 03, 2020
Figure 1 for Single-Shot Pose Estimation of Surgical Robot Instruments' Shafts from Monocular Endoscopic Images
Figure 2 for Single-Shot Pose Estimation of Surgical Robot Instruments' Shafts from Monocular Endoscopic Images
Figure 3 for Single-Shot Pose Estimation of Surgical Robot Instruments' Shafts from Monocular Endoscopic Images
Figure 4 for Single-Shot Pose Estimation of Surgical Robot Instruments' Shafts from Monocular Endoscopic Images
Viaarxiv icon