PaperHub
影响力指数
89.03/100
前 0.6%
全站排名 #418
发表论文40 篇
平均评分4.9
年均产出13.3 篇/年

Seungryong Kim

Associate Professor@Korea Advanced Institute of Science & Technology·韩国·OpenReview
研究方向

dense correspondence · semantic correspondence · matching · descriptor · cross-modal · multi-modal · spatial invariant CNNs

6.0
22

MATRIX: Mask Track Alignment for Interaction-aware Video Generation

ICLR 2026Poster
通讯
5.0
21

PLOT: Pseudo-Labeling via Object Tracking for Monocular 3D Object Detection

ICLR 2026Rejected
5.0
17

Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation

ICLR 2026Poster
5.0
34

3D Scene Prompting for Scene-Consistent Camera-Controllable Video Generation

ICLR 2026Poster
通讯
5.0
16

Text-Aware Image Restoration with Diffusion Models

ICLR 2026Poster
通讯
5.0
25

A Noise is Worth Diffusion Guidance

ICLR 2026Poster
通讯
4.7
19

Effective Test-Time Scaling of Discrete Diffusion through Iterative Refinement

ICLR 2026Rejected
三作
4.5
22

TAG: Tangential Amplifying Guidance for Hallucination-Resistant Diffusion Sampling

ICLR 2026Rejected
4.0
22

Lookahead Unmasking Elicits Reliable Decoding in Diffusion Language Models

ICLR 2026Rejected
二作
4.0
4

URECA: Unique Region Caption Anything

ICLR 2026Withdrawn
通讯
4.0
15

Lookahead Anchoring: Preserving Character Identity in Audio-Driven Human Animation

ICLR 2026Rejected
4.0
5

Tracking Any Point In Multi-View Videos

ICLR 2026Withdrawn
通讯
3.5
5

Feature Warping-and-Conditioning for Representation-Guided Novel View Synthesis

ICLR 2026Withdrawn
通讯
3.5
6

Visual Representation Alignment for Multimodal Large Language Models

ICLR 2026Withdrawn
通讯
3.3
4

Exploring Conditions for Diffusion models in Robotic Control

ICLR 2026Withdrawn
3.3
4

CORAL: Correspondence Alignment For Controllable Person Image Generation

ICLR 2026Withdrawn
通讯
3.0
5

CAMEO: Correspondence-Attention Alignment for Multi-View Diffusion Models

ICLR 2026Withdrawn
通讯
2.0
5

Video Diffusion Model for Point Tracking

ICLR 2026Withdrawn
通讯
7.8
24

Emergent Temporal Correspondences from Video Diffusion Transformers

NeurIPS 2025Poster
通讯
6.8
22

Active Test-time Vision-Language Navigation

NeurIPS 2025Poster
6.8
27

Seg4Diff: Unveiling Open-Vocabulary Semantic Segmentation in Text-to-Image Diffusion Transformers

NeurIPS 2025Poster
通讯
6.8
26

Where and How to Perturb: On the Design of Perturbation Guidance in Diffusion and Flow Models

NeurIPS 2025Poster
通讯
6.4
28

Enhancing 3D Reconstruction for Dynamic Scenes

NeurIPS 2025Poster
通讯
5.8
12

First Attentions Last: Better Exploiting First Attentions for Efficient Parallel Training

NeurIPS 2025Poster
5.8
41

Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting

ICLR 2025Rejected
通讯
5.5
34

PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting

ICLR 2025Rejected
通讯
5.3
15

Geometry-aware Score Distillation via 3D Consistent Noising and Gradients

ICLR 2025Rejected
通讯
4.6
15

PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis

ICML 2025Poster
通讯
4.0
5

Revisiting Emergent Correspondence from Transformers for Self-supervised Multi-frame Depth Estimation

ICLR 2025Withdrawn
通讯
合作者 (20)