Minsu Cho

65

Papers

411

Total Citations

Papers (65)

Unsupervised Object Discovery and Localization in the Wild: Part-Based Matching With Bottom-Up Region Proposals

Learning to Compose Hypercolumns for Visual Correspondence

Learning Correlation Structures for Vision Transformers

Towards More Practical Group Activity Detection: A New Benchmark and Model

Leveraging 3D Geometric Priors in 2D Rotation Symmetry Detection

Few-Shot Pattern Detection via Template Matching and Regression

Axis-level Symmetry Detection with Group-Equivariant Representation

MoReVQA: Exploring Modular Reasoning Models for Video Question Answering

Enhancing 3D Fidelity of Text-to-3D using Cross-View Correspondences

Contrastive Mean-Shift Learning for Generalized Category Discovery

Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform

3D Geometric Shape Assembly via Efficient Point Cloud Matching

Locality-Aware Zero-Shot Human-Object Interaction Detection

Proposal Flow

Thin-Slicing for Pose: Learning to Understand Pose Without Explicit Pose Estimation

Multi-Object Tracking With Quadruplet Convolutional Neural Networks

Deep Metric Learning Beyond Binary Supervision

Relational Knowledge Distillation

Unsupervised Image Matching and Object Discovery as Optimization

Proxy Anchor Loss for Deep Metric Learning

Local-Global Video-Text Interactions for Temporal Grounding

Embedding Transfer With Label Relaxation for Improved Metric Learning

Convolutional Hough Matching Networks

Self-Supervised Equivariant Learning for Oriented Keypoint Detection

TransforMatcher: Match-to-Match Attention for Semantic Correspondence

Self-Taught Metric Learning Without Labels

Reflection and Rotation Symmetry Detection via Equivariant Learning

Integrative Few-Shot Learning for Classification and Segmentation

Fast Point Transformer

Detector-Free Weakly Supervised Group Activity Recognition

Autoregressive Image Generation Using Residual Quantization

Future Transformer for Long-Term Action Anticipation

Learning Rotation-Equivariant Features for Visual Correspondence

Relational Context Learning for Human-Object Interaction Detection

Generalizable Implicit Neural Representations via Instance Pattern Composers

Devil's on the Edges: Selective Quad Attention for Scene Graph Generation

Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation

Soft-Landing Strategy for Alleviating the Task Discrepancy Problem in Temporal Action Localization Tasks

Unsupervised Object Discovery and Tracking in Video Collections

SCNet: Learning Semantic Correspondence

Hyperpixel Flow: Semantic Correspondence With Multi-Layer Neural Features

Deep Hough Voting for Robust Global Registration

Learning Self-Similarity in Space and Time As Generalized Motion for Video Action Recognition

Hypercorrelation Squeeze for Few-Shot Segmentation

Relational Embedding for Few-Shot Classification

Learning To Discover Reflection Symmetry via Polar Matching Convolution

Self-Calibrating Neural Radiance Fields

MotionSqueeze: Neural Motion Feature Learning for Video Understanding

Towards Sequence-Level Training for Visual Tracking

Robust Image Filtering Using Joint Static and Dynamic Guidance

Video Summarization with Large Language Models

Mosaic3D: Foundation Dataset and Model for Open-Vocabulary 3D Segmentation

Combinative Matching for Geometric Shape Assembly

Generic Event Boundary Detection via Denoising Diffusion

Mining GOLD Samples for Conditional GANs

CircleGAN: Generative Adversarial Learning across Spherical Circles

Brick-by-Brick: Combinatorial Construction with Deep Reinforcement Learning

Relational Self-Attention: What's Missing in Attention for Video Understanding

Differentiable Spline Approximations

Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training

PeRFception: Perception using Radiance Fields

Draft-and-Revise: Effective Image Generation with Contextual RQ-Transformer

Peripheral Vision Transformer

Locality-Aware Generalizable Implicit Neural Representation

Activity Grammars for Temporal Action Segmentation