ICLR 2024 accepted papers on Detection and segmentation
42 of the 2,259 papers accepted at ICLR 2024 (1.9%) are on Detection and segmentation, compared with 3.4% at ICLR 2023. They fall into 4 topics.
See ICLR 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Open-set and few-shot detection · 23 papers
- Personalize Segment Anything Model with One Shot (65 citations)
- AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection (32 citations)
- CLIPSelf: Vision Transformer Distills Itself for Open-Vocabulary Dense Prediction (12 citations)
- Convolution Meets LoRA: Parameter Efficient Finetuning for Segment Anything Model (12 citations)
- Decoding Natural Images from EEG for Object Recognition (8 citations)
- ValUES: A Framework for Systematic Validation of Uncertainty Estimation in Semantic Segmentation (4 citations)
- LLMs Meet VLMs: Boost Open Vocabulary Object Detection with Fine-grained Descriptors (3 citations)
- InstructDET: Diversifying Referring Object Detection with Generalized Instructions (2 citations)
- R-MAE: Regions Meet Masked Autoencoders (1 citation)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (1 citation)
- CrIBo: Self-Supervised Learning via Cross-Image Object-Level Bootstrapping
- Cycle Consistency Driven Object Discovery
and 11 more in the interactive map.
CNN features · 12 papers
- FasterViT: Fast Vision Transformers with Hierarchical Attention (35 citations)
- Strategic Preys Make Acute Predators: Enhancing Camouflaged Object Detectors by Generating Camouflaged Objects (29 citations)
- MogaNet: Multi-order Gated Aggregation Network (24 citations)
- DFormer: Rethinking RGBD Representation Learning for Semantic Segmentation (16 citations)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (10 citations)
- FeatUp: A Model-Agnostic Framework for Features at Any Resolution (9 citations)
- CAMIL: Context-Aware Multiple Instance Learning for Cancer Detection and Subtyping in Whole Slide Images (2 citations)
- Bounding Box Stability against Feature Dropout Reflects Detector Generalization across Environments
- Mathematical Justification of Hard Negative Mining via Isometric Approximation Theorem
- Multi-Scale Representations by Varying Window Attention for Semantic Segmentation
- PBADet: A One-Stage Anchor-Free Approach for Part-Body Association
- When Semantic Segmentation Meets Frequency Aliasing
3D detection and driving · 4 papers
- V-DETR: DETR with Vertex Relative Position Encoding for 3D Object Detection (5 citations)
- TopoMLP: A Simple yet Strong Pipeline for Driving Topology Reasoning (2 citations)
- MixSup: Mixed-grained Supervision for Label-efficient LiDAR-based 3D Object Detection (1 citation)
- CALICO: Self-Supervised Camera-LiDAR Contrastive Pre-training for BEV Perception