CSE5519 Advances in Computer Vision

Here is a list of topics we will cover in this course:

Topics and Papers

CSE 5519 (Fall 2025)

Topic A: Semantic Segmentation

A.1 2021 and before

• 🌟 [DeepLabv3+] Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation (ECCV 2018)
• 🌟 [SETR] Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers (CVPR 2021)
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows (CVPR 2021)
• SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers (NeurIPS 2021)

A.2 2022

• [Mask2Former] Masked-attention Mask Transformer for Universal Image Segmentation (CVPR 2022)

A.3 2023 - 2024

• [SAM] 🌟 Segment Anything (arXiv; Apr 5, 2023)
• Segment Anything in High Quality (NeurIPS 2023)
• OMG-Seg: Is One Model Good Enough For All Segmentation? (CVPR 2024)
• SAM 2: Segment Anything in Images and Videos (arXiv; Aug 1, 2024)

A.4 2025

• Grounded SAM v1 and v2
• Dual Semantic Guidance for Open Vocabulary Semantic Segmentation (CVPR 2025)
• Prior2Former: Evidential Modeling of Mask Transformers for Assumption-Free Open-World Panoptic Segmentation (ICCV 2025)

Topic B: Vision-Language Models

B.1 2021 and before

• Learning Visual Features from Large Weakly Supervised Data (ECCV 2016)
• Self-Supervised Learning of Visual Features Through Embedding Images Into Text Topic Spaces (CVPR 2017)
• 🌟 [CLIP] Learning Transferable Visual Models From Natural Language Supervision (arXiv; Feb 26, 2021)

B.2 2022

• Flamingo: a Visual Language Model for Few-Shot Learning (NeurIPS 2022)
• CoCa: Contrastive Captioners are Image-Text Foundation Models (arXiv; May 4, 2022)
• BLIP: Bootstrapping Language-Image Pre-training (ICML 2022)

B.3 2023

• BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models (ICML 2023)
• Visual Instruction Tuning (NeurIPS 2023)
• 🌟 InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning (NeurIPS 2023)
• [Kosmos-1] Language Is Not All You Need – Aligning Perception with Language Models (ICLR 2023)

B.4 2024

• [LLaVA-1.5] Improved Baselines with Visual Instruction Tuning (CVPR 2024)
• Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context (arXiv; Mar 8, 2024)

B.5 2025

• Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models (CVPR 2025)
• ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos (CVPR 2025)

Topic C: Neural Rendering

C.1 2021 and before

• NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (ECCV 2020)
• NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections (CVPR 2021)
• D-NeRF: Neural Radiance Fields for Dynamic Scenes (CVPR 2021)

C.2 2022

• 🌟 Plenoxels: Radiance Fields without Neural Networks (CVPR 2022)
• Mega-nerf: Scalable construction of large-scale nerfs for virtual fly-throughs (CVPR 2022)
• Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022)
• Block-NeRF: Scalable Large Scene Neural View Synthesis (CVPR 2022)
• Depth-Supervised NeRF: Fewer Views and Faster Training for Free (CVPR 2022)

C.3 2023

• 🌟 3D Gaussian Splatting for Real-Time Radiance Field Rendering (SIGGRAPH 2023)
• Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Models (ICCV 2023)
• NoPe-nerf: Optimising neural radiance field with no pose prior (CVPR 2023)

C.4 2024 - 2025

• 🌟 COLMAP-Free 3D Gaussian Splatting (CVPR 2024)
• Flying with Photons: Rendering Novel Views of Propagating Light (ECCV 2024)
• Neural Inverse Rendering from Propagating Light (CVPR 2025)
• 3D Student Splatting and Scooping (CVPR 2025)

Topic D: Image and Video Generation

D.1 2021 and before

• AttnGAN: Fine-Grained Text to Image Generation With Attentional Generative Adversarial Networks (CVPR 2019)
• DF-GAN: A Simple and Effective Baseline for Text-to-Image Synthesis (arXiv; Aug 13, 2020; also CVPR 2022)
• Zero-Shot Text-to-Image Generation (ICML 2021)
• High-Resolution Image Synthesis with Latent Diffusion Models (arXiv; Dec 20, 2021; also CVPR 2022)

D.2 2022

• Make-A-Video: Text-to-Video Generation without Text-Video Data (arXiv; Sep 29, 2022; 2023)
• An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion (arXiv; Aug 2, 2022)

D.3 2023

• DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation (CVPR 2023)
• GLIGEN: Open-Set Grounded Text-to-Image Generation (CVPR 2023)
• InstructPix2Pix: Learning To Follow Image Editing Instructions (CVPR 2023)
• Scalable Diffusion Models with Transformers (ICCV 2023)

D.4 2024

• VideoPoet: A Large Language Model for Zero-Shot Video Generation (ICML 2024)
• MoMA: Multimodal LLM Adapter for Fast Personalized Image Generation (ECCV 2024)
• Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation (arXiv; Jun 10, 2024)

D.5 2025

• There are many works on image and video generation this year that are purely being shown by press releases; this could be an opportunity to demonstrate the state of the art rather than focusing on a specific paper.
• Wan: Open and Advanced Large-Scale Video Generative Models (arXiv; Mar 26, 2025)
  [This is a very long paper, but it is one of the best open source models]

Topic E: Deep Learning for Geometric Computer Vision

E.1 2021 and before

• PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization (ICCV 2015)
• Unsupervised Learning of Depth and Ego-Motion From Video (CVPR 2017)
• GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose (CVPR 2018)

E.2 2022

• MeshLoc: Mesh-Based Visual Localization (ECCV 2022)
• DFNet: Enhance Absolute Pose Regression with Direct Feature Matching (ECCV 2022)
• Map-free Visual Relocalization: Metric Pose Relative to a Single Image (ECCV 2022)

E.3 2023

• Guided Depth Super-Resolution by Deep Anisotropic Diffusion (CVPR 2023)

E.4 2024

• 🌟 DUSt3R: Geometric 3D Vision Made Easy (CVPR 2024)
• Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences (CVPR 2024)
• Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024) and Depth Anything V2 (arXiv; Jun 13, 2024)
• GeoCalib: Learning Single-image Calibration with Geometric Optimization (ECCV 2024)

E.5 2025

• 🌟 VGGT: Visual Geometry Grounded Transformer (CVPR 2025)
• Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass (CVPR 2025)
• Prompting Depth Anything for 4K Resolution Accurate Metric Depth Estimation (CVPR 2025)
• Video Depth Anything: Consistent Depth Estimation for Super-Long Videos (CVPR 2025)

Topic F: Representation Learning

F.1 2021 and before

• [SimCLR] A Simple Framework for Contrastive Learning of Visual Representations (ICML 2020)
• [MoCo] Momentum Contrast for Unsupervised Visual Representation Learning (CVPR 2020)
• Barlow Twins: Self-supervised learning via redundancy reduction (ICML 2021)
• Exploring Simple Siamese Representation Learning (CVPR 2021)

F.2 2022

• [MAE] Masked Autoencoders Are Scalable Vision Learners (CVPR 2022)
• Masked Siamese Networks for Label-Efficient Learning (ECCV 2022)

F.3 2023

• 🌟 [JEPA] Self-supervised learning from images with a joint-embedding predictive architecture (CVPR 2023)
• StableRep: Synthetic Images from Text-to-Image Models Make Strong Visual Representation Learners (NeurIPS 2023)

F.4 2024

• DINOv2: Learning Robust Visual Features without Supervision (TMLR; Jan 2024)
• Long-clip: Unlocking the long-text capability of CLIP (ECCV 2024)

F.5 2025

• FLAIR: VLM with Fine-grained Language-informed Image Representations (CVPR 2025)
• Can Generative Models Improve Self-Supervised Representation Learning? (AAAI 2025)

Topic G: Correspondence Estimation and Structure from Motion

G.1 2022 and before

• Structure-from-Motion Revisited (CVPR 2016)
• SuperGlue: Learning Feature Matching With Graph Neural Networks (CVPR 2020)
• RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (ECCV 2020)
• 🌟 Pixel-Perfect Structure-from-Motion with Featuremetric Refinement (ICCV 2021)
• LoFTR: Detector-Free Local Feature Matching with Transformers (CVPR 2021)

G.2 2023

• LightGlue: Local Feature Matching at Light Speed (ICCV 2023)
• 🌟 Detector-Free Structure from Motion (IMC 2023 → CVPR 2024)

G.3 2024

• RoMa: Robust Dense Feature Matching (CVPR 2024)
• 🌟 Global Structure-from-Motion Revisited (ECCV 2024)

G.4 2025

• MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videos (CVPR 2025)

Topic H: Safety, Robustness & Evaluation of CV Models

H.1 2022 and before

• 🌟 Does Object Recognition Work for Everyone? (CVPR Workshop 2019)
• OccamNets: Mitigating Dataset Bias by Favoring Simpler Hypotheses (ECCV 2022)

H.2 2023

• 🌟 GeoNet: Benchmarking Unsupervised Adaptation across Geographies (CVPR 2023)
• Inspecting the Geographical Representativeness of Images from Text-to-Image Models (ICCV 2023)

H.3 2024

• Concept Arithmetics for Circumventing Concept Inhibition in Diffusion Models (ECCV 2024)
• Efficient Bias Mitigation Without Privileged Information (ECCV 2024)
• Unified Concept Editing in Diffusion Models (WACV 2024)

H.4 2025

• Rethinking Training for De-biasing Text-to-Image Generation: Unlocking the Potential of Stable Diffusion (CVPR 2025)

Topic I: Embodied Computer Vision and Robotics

I.1 2021 and before

• ViNG: Learning Open-World Navigation with Visual Goals (ICRA 2021)
• Pathdreamer: A World Model for Indoor Navigation (ICCV 2021)

I.2 2022

• 🌟 ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints (RSS 2022)
• DayDreamer: World Models for Physical Robot Learning (CoRL 2022)

I.3 2023 - 2024

• GNM: A General Navigation Model to Drive Any Robot (ICRA 2023)
• NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration (ICRA 2024)

I.4 2025

• Navigation World Models (CVPR 2025)

Topic J: Open-Vocabulary Object Detection

J.1 2021 and before

• Open-Vocabulary Object Detection Using Captions (CVPR 2021)
• 🌟 MDETR: Modulated Detection for End-to-End Multi-Modal Understanding (ICCV 2021)

J.2 2022

• Open-vocabulary object detection via vision and language knowledge distillation (ICLR 2022)
• Class-agnostic object detection with multimodal transformer (ECCV 2022)

J.3 2023 - 2024

• CORA: Adapting CLIP for Open-Vocabulary Detection with Region Prompting and Anchor Pre-Matching (CVPR 2023)
• Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection (ECCV 2024)

J.4 2025

• OW-OVD: Unified Open World and Open Vocabulary Object Detection (CVPR 2025)
• Open-World Objectness Modeling Unifies Novel Object Detection (CVPR 2025)
• DINO-X: A Unified Vision Model for Open-World Object Detection and Understanding (arXiv; May 15, 2025)