ArXiv Computer Vision research for Wednesday, June 05, 2024.

00:20: Adapter-X: A Novel General Parameter-Efficient Fine-Tuning Framework for Vision

02:03: A-Bench: Are LMMs Masters at Evaluating AI-generated Images?

03:42: Exploiting LMM-based knowledge for image classification tasks

04:37: EgoSurgery-Tool: A Dataset of Surgical Tool and Hand Detection from Egocentric Open Surgery Videos

06:09: EpidermaQuant: Unsupervised detection and quantification of epidermal differentiation markers on H-DAB-stained images of reconstructed human epidermis

08:15: Enhancing 3D Lane Detection and Topology Reasoning with 2D Lane Priors

09:24: VQUNet: Vector Quantization U-Net for Defending Adversarial Atacks by Regularizing Unwanted Noise

10:36: Enhanced Automotive Object Detection via RGB-D Fusion in a DiffusionDet Framework

11:42: ZeroPur: Succinct Training-Free Adversarial Purification

13:23: Tiny models from tiny data: Textual and null-text inversion for few-shot distillation

15:10: Multi-Task Multi-Scale Contrastive Knowledge Distillation for Efficient Medical Image Segmentation

16:44: Dynamic 3D Gaussian Fields for Urban Areas

18:10: MMCL: Boosting Deformable DETR-Based Detectors with Multi-Class Min-Margin Contrastive Learning for Superior Prohibited Item Detection

20:02: FAPNet: An Effective Frequency Adaptive Point-based Eye Tracker

21:52: Ouroboros3D: Image-to-3D Generation via 3D-aware Recursive Diffusion

23:14: Situation Monitor: Diversity-Driven Zero-Shot Out-of-Distribution Detection using Budding Ensemble Architecture for Object Detection

24:28: Writing Order Recovery in Complex and Long Static Handwriting

25:50: Identification of Stone Deterioration Patterns with Large Multimodal Models

26:58: Searching Priors Makes Text-to-Video Synthesis Better

28:32: Interactive Image Selection and Training for Brain Tumor Segmentation Network

29:35: Global Clipper: Enhancing Safety and Reliability of Transformer-based Object Detection Models

30:53: Generative Diffusion Models for Fast Simulations of Particle Collisions at CERN

31:52: Prompt-based Visual Alignment for Zero-shot Policy Transfer

33:33: ADer: A Comprehensive Benchmark for Multi-class Visual Anomaly Detection