AI Paper Atlas

Research map / Benchmarks / UCF101 split 1

UCF101 split 1 leaderboard

23 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on UCF101 split 1. Best accuracy: 98.8 (BIKE ViT-L, CVPR 2023).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

UCF101 split 1 · accuracy

23 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
198.8BIKE ViT-L
CLIP ViT-L
Bidirectional Cross-Modal Knowledge Exploration for Video Recognition With Pre-Trained Vision-Language Models
CVPR 2023 · Table A
CLIP and zero-shot
298Two-Stream I3D, Full-FT
Inception-v1 + Kinetics
Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset
CVPR 2017 · Feature transfer table
Action recognition
397.5DualFormer-S (4x3 views)
Kinetics-400 pretrain
DualFormer: Local-Global Stratified Transformer for Efficient Video Recognition
ECCV 2022 · Table 3
Video segmentation and tracking
496.7VidTr-L
ViT-B
VidTr: Video Transformer Without Convolutions
ICCV 2021 · Table 5
Action recognition
595.5RGB + OFF(RGB) + Flow + OFF(Flow)Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognition
CVPR 2018 · Table: Experimental results fo
Action recognition
695SCALE_ft
rhoBYOL
Spatio-Temporal Crop Aggregation for Video Representation Learning
ICCV 2023 · Table 2
Action recognition
794.2Model D (full pyramid)
BN-Inception
Spatiotemporal Pyramid Network for Video Action Recognition
CVPR 2017 · Table 6
Action recognition
894.1Static App. + Motion + Warped-DiffRecognize Actions by Disentangling Components of Dynamics
CVPR 2018 · Table 6
Action recognition
993.7CO2FI + ASYN
VGG-16
Action Recognition With Coarse-to-Fine Deep Feature Integration and Asynchronous Fusion
AAAI 2018 · Table 2
Action recognition
1093.5BN-Inception+TSN (two-stream)
BN-Inception
Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
ECCV 2016 · Table 5
Action recognition
13 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.