论文阅读笔记

按时间倒序排列的论文阅读记录。

格式：发表日期 标题 — tag1 tag2

2026

• 05-24 TinyFormer: YOLO-DETR Hybrid Tiny Object Detection — yolo detr tiny-object real-time
• 05-26 LocateAnything: Parallel Box Decoding for VLM Grounding — open-vocabulary visual-grounding parallel-decoding VLM
• 05-10 AnyDepth: Any-Depth Object Detection — adaptive-inference RT-DETR YOLOv12 cvpr2026
• 05-10 HSGDet: Open World Detection in Remote Sensing — open-world remote-sensing unknown-label cvpr2026
• 05-23 DisDop: Domain Prior Distillation for OVAOD — open-vocabulary aerial-detection distillation
• 05-16 CANSURF: ASV Surface Debris Detection — yolo object-detection marine-debris dataset
• 04-22 CPL++: Weakly Supervised Visual Grounding — visual-grounding weakly-supervised pseudo-label
• 04-22 Falcon Perception — vlm open-vocabulary-segmentation early-fusion
• 04-18 FAA: Fourier Angle Alignment — oriented-detection remote-sensing fourier cvpr2026
• 04-16 SkyFind: UAV REC Benchmark — uav rec dataset
• 04-12 Seg2Change: Open-Vocabulary Change Detection — remote-sensing change-detection open-vocabulary
• 03-19 EdgeCrafter: Compact ViT Edge Dense Prediction — ViT edge distillation object-detection pose segmentation
• 03-15 UAVBench & UAVIT-1M: Low-Altitude UAV VL Benchmark — MLLM low-altitude-UAV benchmark instruction-tuning
• 02-17 YOLO26: A Comprehensive Architecture Overview — yolo26 architecture
• 02-03 YOLOE-26: Open-Vocabulary Segmentation — open-vocabulary segmentation
• 01-31 VLDet: Open-Vocabulary Detection via Visual-Language Alignment — open-vocabulary CLIP feature-pyramid yolo-world
• 01-19 YOLO26: NMS-Free End-to-End Framework — yolo nms-free

2025

• 11-24 RefDrone: REC Benchmark for Drone Scenes — rec uav benchmark
• 11-15 RF-DETR: Neural Architecture Search — DETR NAS
• 10-17 ReCon: Data Augmentation with Rectification and Alignment — neurips2025 data-augmentation diffusion
• 10-14 Rex-Omni: Detect Anything via Next Point Prediction — cvpr2026 mllm object-detection
• 09-25 DEIMv2: Real-Time Object Detection Meets DINOv3 — dinov3 deim real-time-detection transformer
• 04-29 CoDrone: UAV Oriented Detection Benchmark — uav obb
• 03-10 YOLOE: Real-Time Seeing Anything — open-vocabulary YOLO visual-prompt prompt-free re-parameterization

2024

• 12-13 RemDet: Efficient Model Design for UAV Detection — uav-detection efficient-design
• 08-17 LAE: Locate Anything on Earth for RS OVD — open-vocabulary remote-sensing DINO data-engine
• 01-30 YOLO-World: Real-Time Open-Vocabulary Object Detection — open-vocabulary YOLO real-time vision-language
• 09-25 YOLO26 Jetson Benchmark — benchmark
• 07-23 ESOD: Efficient Small Object Detection on High-Resolution Images — small-object high-resolution sparse-detection
• 04-29 CoDrone: UAV Oriented Detection Benchmark — uav obb

2023

• 11-20 CastDet: CLIP-Activated OVAOD — open-vocabulary aerial-detection student-teacher CLIP
• 06-19 RemoteCLIP: VL Foundation Model for RS — vision-language remote-sensing CLIP data-scaling
• 03-16 Large Selective Kernel Network — remote-sensing backbone
• 03-09 Grounding DINO: Marrying DINO with Grounded Pre-Training — open-vocabulary grounding DINO vision-language

2021

• 11-23 FGD: Focal/Gobal Distillation for Detectors — knowledge-distillation cvpr2022

2020

• 01-17 Detection and Tracking Meet Drones Challenge — drone detection
• 01-01 DIOR Dataset — remote-sensing dataset