Semantic Saliency from Multi-Modal Large Language Model Scene Understanding Maps
cs.CV
Submitted: 2026-10-11
Updated: 2026-10-11
Code: https://github.com/starsky77/AUTO-SUM-Source-Code
Sources
- Saliency Prediction in the Deep Learning Era: Successes, Limitations, and Future Challenges
- DeepFix: A Fully Convolutional Neural Network for predicting Human Eye Fixations
- SalGAN: Visual Saliency Prediction with Generative Adversarial Networks
- Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models
- GPT-4 Technical Report
- Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
- Ovis: Structural Embedding Alignment for Multimodal Large Language Model
- ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
- Deep Gaze I: Boosting Saliency Prediction with Feature Maps Trained on ImageNet
- CoCa: Contrastive Captioners are Image-Text Foundation Models
- SAM 3: Segment Anything with Concepts
- Jasper and Stella: distillation of SOTA embedding models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models