Object Detection
Object detection is a crucial task in the field of computer vision, aimed at identifying and locating specific objects within images. This task involves automatically detecting target objects in an image and returning their position and category information, which has significant application value in areas such as autonomous driving, security surveillance, and medical image analysis. The development of object detection technology has significantly enhanced machines' ability to understand complex environments, providing critical support for intelligent systems.
RL [10] Lpixel
DNTR
Cascade RCNN (ConvNext-T, RAW pre-training)
Aquavision
hybrid incremental net
Cascade R-CNN (R50-FPN)
LeYOLO-Nano
MOAT-3 22K+1K
UniRepLKNet-XL++
Mr. DETR (Swin-L, 1x, 4scale)
RepPoints + Self-adaptation
Co-DETR
EVA
TridentNet
S-RCNN+Ours
IterDet (Faster RCNN, ResNet50, 2 iterations)
YOLOv5
GLIP-T
EfficientDet-D2
PNe within NGC1380 & NGC1404
Logo-Yolo
MiPa
ERGO-12
iGDINO MAC-SORT
KnowZRel
YOLOv8x
hybrid incremental net
YoloV8
PANet++
Co-DETR (single-scale)
YOLOX-L
PP-PicoDet-L
TarDAL
Mask2Former (R50)
BIRANet(RGB+Radar)
Grounding DINO 1.5 Pro
UGainS
Attention-based Joint Detection of Object and Semantic Part
TinyissimoYOLO-v8
DETReg (ours)
YOLOv7+Inner-IoU
Faster R-CNN
SynCo (ResNet-50) 200ep
MILA
CDDMSL
CAFR
MS-PAD
Relation-DETR (ResNet50 1x)
PGD-YOLOv8
Synth Pretrained Faster R-CNN ResNeXt-101-FPN
YOLOv8+SCALE
YOLO
Optim [39] Lpixel
Geo-trax
YOLT
MMPedestron
CDSSD
Swin-T (ImageNet-1k pretrain)
DetectoRS + LAEM
VSTAM
FFAVOD-SpotNet with U-Net
UniverseNet-20.08
GHOST
SSOD + Crop (L + U)
CZ Det.
YOLOv8-M
UniverseNet
YOLOX-L
LeapMotor_Det
IterDet (Faster RCNN, ResNet50, 2 iterations)