Picture for Masanori Suganuma

Masanori Suganuma

An Improved Method for Personalizing Diffusion Models

Add code
Jul 07, 2024
Viaarxiv icon

SBCFormer: Lightweight Network Capable of Full-size ImageNet Classification at 1 FPS on Single Board Computers

Add code
Nov 07, 2023
Viaarxiv icon

Visual Abductive Reasoning Meets Driving Hazard Prediction: Problem Formulation and Dataset

Add code
Oct 10, 2023
Viaarxiv icon

Contextual Affinity Distillation for Image Anomaly Detection

Add code
Jul 06, 2023
Figure 1 for Contextual Affinity Distillation for Image Anomaly Detection
Figure 2 for Contextual Affinity Distillation for Image Anomaly Detection
Figure 3 for Contextual Affinity Distillation for Image Anomaly Detection
Figure 4 for Contextual Affinity Distillation for Image Anomaly Detection
Viaarxiv icon

That's BAD: Blind Anomaly Detection by Implicit Local Feature Clustering

Add code
Jul 06, 2023
Figure 1 for That's BAD: Blind Anomaly Detection by Implicit Local Feature Clustering
Figure 2 for That's BAD: Blind Anomaly Detection by Implicit Local Feature Clustering
Figure 3 for That's BAD: Blind Anomaly Detection by Implicit Local Feature Clustering
Figure 4 for That's BAD: Blind Anomaly Detection by Implicit Local Feature Clustering
Viaarxiv icon

Reference-based Motion Blur Removal: Learning to Utilize Sharpness in the Reference Image

Add code
Jul 06, 2023
Figure 1 for Reference-based Motion Blur Removal: Learning to Utilize Sharpness in the Reference Image
Figure 2 for Reference-based Motion Blur Removal: Learning to Utilize Sharpness in the Reference Image
Figure 3 for Reference-based Motion Blur Removal: Learning to Utilize Sharpness in the Reference Image
Figure 4 for Reference-based Motion Blur Removal: Learning to Utilize Sharpness in the Reference Image
Viaarxiv icon

RefVSR++: Exploiting Reference Inputs for Reference-based Video Super-resolution

Add code
Jul 06, 2023
Figure 1 for RefVSR++: Exploiting Reference Inputs for Reference-based Video Super-resolution
Figure 2 for RefVSR++: Exploiting Reference Inputs for Reference-based Video Super-resolution
Figure 3 for RefVSR++: Exploiting Reference Inputs for Reference-based Video Super-resolution
Figure 4 for RefVSR++: Exploiting Reference Inputs for Reference-based Video Super-resolution
Viaarxiv icon

More Practical Scenario of Open-set Object Detection: Open at Category Level and Closed at Super-category Level

Add code
Jul 20, 2022
Figure 1 for More Practical Scenario of Open-set Object Detection: Open at Category Level and Closed at Super-category Level
Figure 2 for More Practical Scenario of Open-set Object Detection: Open at Category Level and Closed at Super-category Level
Figure 3 for More Practical Scenario of Open-set Object Detection: Open at Category Level and Closed at Super-category Level
Figure 4 for More Practical Scenario of Open-set Object Detection: Open at Category Level and Closed at Super-category Level
Viaarxiv icon

GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features

Add code
Jul 20, 2022
Figure 1 for GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features
Figure 2 for GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features
Figure 3 for GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features
Figure 4 for GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features
Viaarxiv icon

Single-image Defocus Deblurring by Integration of Defocus Map Prediction Tracing the Inverse Problem Computation

Add code
Jul 07, 2022
Figure 1 for Single-image Defocus Deblurring by Integration of Defocus Map Prediction Tracing the Inverse Problem Computation
Figure 2 for Single-image Defocus Deblurring by Integration of Defocus Map Prediction Tracing the Inverse Problem Computation
Figure 3 for Single-image Defocus Deblurring by Integration of Defocus Map Prediction Tracing the Inverse Problem Computation
Figure 4 for Single-image Defocus Deblurring by Integration of Defocus Map Prediction Tracing the Inverse Problem Computation
Viaarxiv icon