Video Retrieval On Didemo

評価指標

text-to-video R@1

text-to-video R@10

text-to-video R@5

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	text-to-video R@1	text-to-video R@10	text-to-video R@5	Paper Title	Repository
RTQ	57.6	89.9	84.1	RTQ: Rethinking Video-language Understanding Based on Image-text Model
FROZEN	31.0	72.4	59.8	Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval
DiffusionRet+QB-Norm	48.9	83.3	75.5	DiffusionRet: Generative Text-Video Retrieval with Diffusion Model
HunYuan_tvr (huge)	52.7	85.2	77.8	Tencent Text-Video Retrieval: Hierarchical Cross-Modal Interactions with Multi-Level Representations	-
HD-VILA	28.8	69.1	57.4	Advancing High-Resolution Video-Language Representation with Large-Scale Video Transcriptions
QB-Norm+CLIP4Clip	43.5	80.9	71.4	Cross Modal Retrieval with Querybank Normalisation
Cap4Video	52.0	87.5	79.4	Cap4Video: What Can Auxiliary Captions Do for Text-Video Retrieval?
STAN	54.6	85.1	78.4	Revisiting Temporal Modeling for CLIP-based Image-to-Video Knowledge Transferring
CLIP-ViP	55.3	89.3	82	CLIP-ViP: Adapting Pre-trained Image-Text Model to Video-Language Representation Alignment
ALPRO	35.9	78.8	67.5	Align and Prompt: Video-and-Language Pre-training with Entity Prompts
DRL	49.0	84.5	76.5	Disentangled Representation Learning for Text-Video Retrieval
VAST	72.0	91.4	89.0	VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset
DMAE (ViT-B/32)	52.7	86.6	79.3	Dual-Modal Attention-Enhanced Text-Video Retrieval with Triplet Partial Margin Contrastive Learning	-
mPLUG-2	56.4	85.2	79.1	mPLUG-2: A Modularized Multi-modal Foundation Model Across Text, Image and Video
Clover	50.1	85.6	76.7	Clover: Towards A Unified Video-Language Alignment and Fusion Model
VindLU	61.2	91.0	85.8	VindLU: A Recipe for Effective Video-and-Language Pretraining
Collaborative Experts	16.1	54.4	41.1	Use What You Have: Video Retrieval Using Representations From Collaborative Experts
MuLTI	56.5	87.0	80.2	MuLTI: Efficient Video-and-Language Understanding with Text-Guided MultiWay-Sampler and Multiple Choice Modeling	-
UMT-L (ViT-L/16)	70.4	93.5	90.1	Unmasked Teacher: Towards Training-Efficient Video Foundation Models
HiTeA	56.5	89.7	81.7	HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training	-

0 of 40 row(s) selected.