AI Paper Atlas

Research map / Benchmarks / A2D-Sentences

A2D-Sentences leaderboard

18 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on A2D-Sentences. Best mAP: 61.4 (VD-IT (Video Diffusion), ECCV 2024).

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.

A2D-Sentences · mAP

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

#mAP ↑MethodPaperTopic
161.4VD-IT (Video Diffusion)
ModelScopeT2V
Exploring Pre-trained Text-to-Video Diffusion Models for Referring Video Object Segmentation
ECCV 2024 · Table 2
Video segmentation and tracking
261.2SPOT-7BSPOT: Spatiotemporal Prompt Optimization for Motion-Stabilized MLLM-Guided Video Segmentation
CVPR 2026 · Table 2
Video-language understanding
359.9LoSh-S (w = 5)
Video-Swin-B
LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation
CVPR 2024 · Table 1
Video segmentation and tracking
458.5SgMg
Video-Swin-B
Spectrum-guided Multi-granularity Referring Video Object Segmentation
ICCV 2023 · Table 2
Video segmentation and tracking
557.3SOC (image pretrain)
Video-Swin-B
SOC: Semantic-Assisted Object Cluster for Referring Video Object Segmentation
NeurIPS 2023 · Table 2
Video-language understanding
657.2DsHmp
Video-Swin-Tiny
Decoupling Static and Hierarchical Motion Perception for Referring Video Segmentation
CVPR 2024 · Table 5
Video segmentation and tracking
756.7HTML
Video-Swin-B
HTML: Hybrid Temporal-scale Multimodal Learning Framework for Referring Video Object Segmentation
ICCV 2023 · Table 3
Video segmentation and tracking
855ReferFormer
Video-Swin-B
Language As Queries for Referring Video Object Segmentation
CVPR 2022 · Table 2
CNN features
946.1MTTR (w=10)
Video Swin-T; RoBERTa-base
End-to-End Referring Video Object Segmentation With Multimodal Transformers
CVPR 2022 · Table 1
Video-language understanding
1042.9Dr2Net
Video Swin-tiny
Dr2Net: Dynamic Reversible Dual-Residual Networks for Memory-Efficient Finetuning
CVPR 2024 · Table 2
RNNs and normalization

A2D-Sentences · mIoU

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

#mIoU ↑MethodPaperTopic
173.6ReferDINO
Swin-B
ReferDINO: Referring Video Object Segmentation with Visual Grounding Foundations
ICCV 2025 · Table 1
Video segmentation and tracking
273.1LoSh-S (w = 5)
Video-Swin-B
LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation
CVPR 2024 · Table 1
Video segmentation and tracking
370.5OnlineRefer (w=5)
Video-Swin-B
OnlineRefer: A Simple Online Baseline for Referring Video Object Segmentation
ICCV 2023 · Table 2
Video segmentation and tracking
468.6TempCD (Ours)
ResNet-50
Temporal Collection and Distribution for Referring Video Object Segmentation
ICCV 2023 · Table 2
Video segmentation and tracking
566.9SOC
Video-Swin-T
SOC: Semantic-Assisted Object Cluster for Referring Video Object Segmentation
NeurIPS 2023 · Table 2
Video-language understanding
656.1OursCollaborative Spatial-Temporal Modeling for Language-Queried Video Actor Segmentation
CVPR 2021 · Table 1
Video segmentation and tracking
753.1CMDyConv (Ours)
I3D
Context Modulated Dynamic Networks for Actor and Action Video Segmentation with Language Queries
AAAI 2020 · Table 1
Action recognition
849Ours (RGB)Asymmetric Cross-Guided Attention Network for Actor and Action Video Segmentation From Natural Language Query
ICCV 2019 · Table 1
Video-language understanding
943.2Ours+CFSA
DeepLab-101
Referring Segmentation in Images and Videos with Cross-Modal Self-Attention Network
TPAMI 2021 · Table 7
Video segmentation and tracking
1042.6RGB + FlowActor and Action Video Segmentation From a Sentence
CVPR 2018 · Table 1
Action recognition

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