SOTAVerified

Object Recognition

Object recognition is a computer vision technique for detecting + classifying objects in images or videos. Since this is a combined task of object detection plus image classification, the state-of-the-art tables are recorded for each component task here and here.

( Image credit: Tensorflow Object Detection API )

Papers

Showing 2650 of 2042 papers

TitleStatusHype
NUDT4MSTAR: A Large Dataset and Benchmark Towards Remote Sensing Object Recognition in the WildCode2
InstructSAM: A Training-Free Framework for Instruction-Oriented Remote Sensing Object RecognitionCode2
HAKE: A Knowledge Engine Foundation for Human Activity UnderstandingCode2
GCNet: Non-local Networks Meet Squeeze-Excitation Networks and BeyondCode2
Learning Transferable Visual Models From Natural Language SupervisionCode2
Hypergraph Neural NetworksCode2
Omni3D: A Large Benchmark and Model for 3D Object Detection in the WildCode2
Discover and Cure: Concept-aware Mitigation of Spurious CorrelationCode1
DetMatch: Two Teachers are Better Than One for Joint 2D and 3D Semi-Supervised Object DetectionCode1
Distributed Deep Neural Networks over the Cloud, the Edge and End DevicesCode1
Describing Textures in the WildCode1
DesCo: Learning Object Recognition with Rich Language DescriptionsCode1
Divergences in Color Perception between Deep Neural Networks and HumansCode1
DeepScores -- A Dataset for Segmentation, Detection and Classification of Tiny ObjectsCode1
Benchmarking Multimodal Mathematical Reasoning with Explicit Visual DependencyCode1
Deep Subdomain Adaptation Network for Image ClassificationCode1
Deep Learning for Event-based Vision: A Comprehensive Survey and BenchmarksCode1
Deep Gaze I: Boosting Saliency Prediction with Feature Maps Trained on ImageNetCode1
Deep Predictive Coding Networks for Video Prediction and Unsupervised LearningCode1
Densely Connected Convolutional NetworksCode1
Do Adversarially Robust ImageNet Models Transfer Better?Code1
Debiased Self-Training for Semi-Supervised LearningCode1
Decoding Natural Images from EEG for Object RecognitionCode1
3D ShapeNets: A Deep Representation for Volumetric ShapesCode1
CREST: An Efficient Conjointly-trained Spike-driven Framework for Event-based Object Detection Exploiting Spatiotemporal DynamicsCode1
Show:102550
← PrevPage 2 of 82Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Imagenshape bias98.7Unverified
2Stable Diffusionshape bias92.7Unverified
3Partishape bias91.7Unverified
4ViT-22B-384shape bias86.4Unverified
5ViT-22B-560shape bias83.8Unverified
6CLIP (ViT-B)shape bias79.9Unverified
7ViT-22B-224shape bias78Unverified
8ResNet-50 (L2 eps 5.0 adv trained)shape bias69.5Unverified
9ResNet-50 (with strong augmentations)shape bias62.2Unverified
10SWSL (ResNeXt-101)shape bias49.8Unverified
#ModelMetricClaimedVerifiedStatus
1Spike-VGG11Accuracy (% )85.55Unverified
2SSNNAccuracy (% )78.57Unverified
#ModelMetricClaimedVerifiedStatus
1Spike-VGG11Accuracy (% )85.62Unverified
2SSNNAccuracy (% )79.25Unverified
#ModelMetricClaimedVerifiedStatus
1ObjectNet-BaselineTop 5 Accuracy18.75Unverified
2yunTop 5 Accuracy14.75Unverified
#ModelMetricClaimedVerifiedStatus
1ObjectNet-BaselineTop 5 Accuracy52.24Unverified
2DYTop 5 Accuracy0.08Unverified
#ModelMetricClaimedVerifiedStatus
1ObjectNet-BaselineTop 5 Accuracy52.24Unverified
2AJ2021Top 5 Accuracy27.68Unverified
#ModelMetricClaimedVerifiedStatus
1SSNNAccuracy (% )94.91Unverified
#ModelMetricClaimedVerifiedStatus
1Faster-RCNNmAP30.39Unverified
#ModelMetricClaimedVerifiedStatus
1Spike-VGG11Accuracy (% )96Unverified