CVPR 2021 accepted papers on Vision-language and multimodal
83 of the 1,661 papers accepted at CVPR 2021 (5.0%) are on Vision-language and multimodal, compared with 4.4% at CVPR 2020. They fall into 6 topics.
See CVPR 2021 in the interactive map
Submitting to CVPR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Human and embodied modeling · 38 papers
- Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts (615 citations)
- Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition (444 citations)
- Exploring and Distilling Posterior and Prior Knowledge for Radiology Report Generation (346 citations)
- RSTNet: Captioning With Adaptive Attention on Visual and Non-Visual Words (279 citations)
- Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback (244 citations)
- Seeing Out of the Box: End-to-End Pre-Training for Vision-Language Representation Learning (217 citations)
- Causal Attention for Vision-Language Tasks (158 citations)
- TextOCR: Towards Large-Scale End-to-End Reasoning for Arbitrary-Shaped Scene Text (157 citations)
- ArtEmis: Affective Language for Visual Art (146 citations)
- Nutrition5k: Towards Automatic Nutritional Understanding of Generic Food (137 citations)
- SelfDoc: Self-Supervised Document Representation Learning (122 citations)
- Kaleido-BERT: Vision-Language Pre-Training on Fashion Domain (114 citations)
and 26 more in the interactive map.
Cross-modal learning · 29 papers
- Learning the Best Pooling Strategy for Visual Semantic Embedding (284 citations)
- Probabilistic Embeddings for Cross-Modal Retrieval (253 citations)
- Audio-Visual Instance Discrimination with Cross-Modal Agreement (214 citations)
- Progressive Modality Reinforcement for Human Multimodal Emotion Recognition From Unaligned Multimodal Sequences (208 citations)
- T2VLAD: Global-Local Sequence Alignment for Text-Video Retrieval (191 citations)
- Encoder Fusion Network With Co-Attention Embedding for Referring Image Segmentation (179 citations)
- Multimodal Contrastive Training for Visual Representation Learning (172 citations)
- VisualVoice: Audio-Visual Speech Separation With Cross-Modal Consistency (165 citations)
- Cross-Modal Collaborative Representation Learning and a Large-Scale RGBT Benchmark for Crowd Counting (160 citations)
- StyleMeUp: Towards Style-Agnostic Sketch-Based Image Retrieval (112 citations)
- Learning Cross-Modal Retrieval With Noisy Labels (111 citations)
- Semantic Audio-Visual Navigation (90 citations)
and 17 more in the interactive map.
Visual question answering · 9 papers
- Counterfactual VQA: A Cause-Effect Look at Language Bias (420 citations)
- Roses Are Red, Violets Are Blue... but Should VQA Expect Them To? (85 citations)
- Predicting Human Scanpaths in Visual Question Answering (62 citations)
- Perception Matters: Detecting Perception Failures of VQA Models Using Metamorphic Testing (40 citations)
- Separating Skills and Concepts for Novel Visual Question Answering (33 citations)
- Found a Reason for me? Weakly-supervised Grounded Visual Question Answering using Capsules (30 citations)
- How Transferable Are Reasoning Patterns in VQA? (24 citations)
- PQA: Perceptual Question Answering (3 citations)
- Believe It or Not: Counterfactual Bias Manipulation in Visual Question Answering
Robot action and navigation · 5 papers
- VLN BERT: A Recurrent Vision-and-Language BERT for Navigation (246 citations)
- Structured Scene Memory for Vision-Language Navigation (116 citations)
- Topological Planning With Transformers for Vision-and-Language Navigation (106 citations)
- Room-and-Object Aware Knowledge Reasoning for Remote Embodied Referring Expression (82 citations)
- Scene-Intuitive Agent for Remote Embodied Visual Grounding (41 citations)
CLIP and zero-shot · 1 paper
- Open World Compositional Zero-Shot Learning (150 citations)
Multimodal LLMs · 1 paper
- Transformation Driven Visual Reasoning (18 citations)
