Jul 2026
Vision as Unified Multimodal Generation
arXiv ↗We formulate computer vision as unified multimodal generation, where heterogeneous visual tasks are expressed in the native text and image generation spaces of a unified m
💛 40
📚 Awesome Computer Vision
14,912 papers across 0 tags, ranked by community signal and explained.
Jul 2026
Do All Visual Tokens Matter Equally? Object-Evidence Preserving Token Merging for Vision-Language Retrieval
arXiv ↗Multi-vector vision-language retrieval preserves fine-grained visual evidence through maximum-similarity late interaction, but dense image-side tokens make storage and s
💛 16
Jul 2026
MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing
arXiv ↗Recent advances in video diffusion models have enabled either long single-view generation through temporal autoregression, or short multi-view synthesis through bidirectio
💛 8
Jul 2026
SynCity 3000: Bootstrapping Scene-Scale 3D Diffusion
arXiv ↗We present SynCity 3000, a framework for generating 3D scenes that are globally coherent while enabling fine-grained layout control. Building on the ability of current ima
💛 3
Jul 2026
TrajLoc: Trajectory-Attention Localization for Multi-Object Motion Control
arXiv ↗Controlling the motion of multiple objects in image-to-video (I2V) generation requires preserving object identities while enforcing adherence to distinct target trajectori
💛 1
Jul 2026
PixCon: Clean-Positive Contrastive Learning for Foundation-Model Semi-Supervised Segmentation
arXiv ↗Semi-supervised semantic segmentation (SSSS) has long turned on one question, which pseudo-labels to trust, and answered it with ever more careful confidence filtering. Fo
💛 2
Jul 2026
LIST3R: Long-sequence Instance-aware 3D Reconstruction
arXiv ↗We present LIST3R, an instance-aware framework for long-sequence 3D reconstruction inspired by the way humans organize spatial memory around stable and recognizable object
Jul 2026
Pano2World: End-to-End 3D Generation via Unified Multi-View Sequences
arXiv ↗A single panorama captures the full visual sphere from one camera center, yet confines users to looking around in place without enabling true scene exploration. Conver
Jul 2026
World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video
arXiv ↗We present World from Motion, a method for generating freely renderable dynamic 3D Gaussian representations from monocular videos. Our approach conditions a video model on
Jul 2026
Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition
arXiv ↗Urban-scale Visual Place Recognition (VPR) aims to identify the geographic location of a query image by matching it against a geo-tagged database. While recent methods ach
Jul 2026
Segmenting, Fast and Slow: Real-Time Open-Vocabulary Video Instance Segmentation with Dual-Path Processing
arXiv ↗Object-centric models inspired by DETR have become the dominant paradigm for open-vocabulary video instance segmentation (OV-VIS). While recent efforts have reduced the co
Jul 2026
Progressive Pose-Guided 4D Animal Reconstruction from Monocular Video
arXiv ↗Reconstructing 4D animals from monocular videos is challenging due to large inter-species variation, complex articulations, and the lack of reliable templates. Existing ap
Jul 2026
Does Your ViT Still Need U-Net for Segmentation?
arXiv ↗Medical image segmentation is dominated by U-Net-style encoder-decoder architectures. Vision Transformers (ViTs) overcome the limited receptive field of convolutional netw
Jul 2026
MVDGC: Joint 3D and 2D Multi-view Pedestrian Detection via Dual Geometric Constraints
arXiv ↗The core challenge in multi-view pedestrian detection (MVPD) lies in effective aggregation of visual features from different viewpoints for robust occlusion reasoning. Rec
Jul 2026
CORGI: Consistency-Aware 3D Dog Reconstruction from a Single Image in the Wild
arXiv ↗Reconstructing high-fidelity 3D models of highly articulated animals, such as dogs, from a single in-the-wild image remains a formidable challenge. In this paper, we intro
Jul 2026
Personalized Object Identification and Localization via In-Context Inference with Vision-Language Models
arXiv ↗Personalized object localization (POL) localizes an object instance in a query image based on a few reference images with bounding-box annotations and a target object labe
Jul 2026
Learning to Compose: Revisiting Proxy Task Design for Zero-Shot Composed Image Retrieval
arXiv ↗Composed Image Retrieval (CIR) retrieves a target image from a reference image and a textual modification. While supervised CIR relies on costly triplets, Zero-Shot CIR (Z
Jul 2026
Vitality-Aware Compression for Efficient Image-to-Shape Diffusion Transformers
arXiv ↗We propose the first compression approach for image-to-shape Diffusion Transformers (DiTs) that substantially reduces model size while preserving geometric fidelity. Despi
Jul 2026
MedCAGD: Context-Aware Gated Decoder for Efficient Medical Image Segmentation
arXiv ↗Medical image segmentation relies on the ability of encoder-decoder architectures to translate rich feature representations into accurate pixel-level predictions under cha
Jul 2026
Robust 3D Alignment of Generative Reconstructions via Partial Monocular Observations
arXiv ↗Aligning generative 3D reconstructions with partial monocular observations is a critical but under-explored challenge in computer vision. This task is inherently ill-posed
Jul 2026
Restore3D: Breathing Life into Broken Objects with Shape and Texture Restoration
arXiv ↗Restoring incomplete or damaged 3D objects is crucial for cultural heritage preservation, occluded object reconstruction, and artistic design. Existing methods primarily f
Jul 2026
Active Spatial Guidance: Eliminating Injected Positional Mechanisms in Vision Transformers
arXiv ↗Vision Transformers (ViTs) commonly rely on injected positional mechanisms to address self-attention's permutation invariance. Motivated by the spatial regularities of nat
Jul 2026
LUMA: Benchmarking Segmentation via a Lightweight Universal Mask Adapter
arXiv ↗Comparing transformer backbones for image segmentation is confounded: each is paired with a different decoder, recipe, and pretraining, so reported differences rarely refl
Jul 2026
Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation
arXiv ↗Self-Supervised Monocular Depth Estimation (MDE) has garnered attention in recent years due to its independence from ground truth. However, most existing models are limite