Sunghwan Hong

Seokju Cho, Heeseong Shin, Sunghwan Hong et al. · nvidia, utoronto

Open-vocabulary semantic segmentation presents the challenge of labeling each pixel within an image based on a wide range of text descriptions. In this work, we introduce a novel cost-based approach to adapt vision-language foundation models, notably CLIP, for the intricate task of semantic segmentation. Through aggregating the cosine similarity score, i.e., the cost volume between image and text embeddings, our method potently adapts CLIP for segmenting seen and unseen classes by fine-tuning its encoders, addressing the challenges faced by existing methods in handling unseen classes. Building upon this, we explore methods to effectively aggregate the cost volume considering its multi-modal nature of being established between image and text embeddings. Furthermore, we examine various methods for efficiently fine-tuning CLIP.

Sunghwan Hong, Seokju Cho, Jisu Nam et al. · nvidia, utoronto

This paper presents a novel cost aggregation network, called Volumetric Aggregation with Transformers (VAT), for few-shot segmentation. The use of transformers can benefit correlation map aggregation through self-attention over a global receptive field. However, the tokenization of a correlation map for transformer processing can be detrimental, because the discontinuity at token boundaries reduces the local context available near the token edges and decreases inductive bias. To address this problem, we propose a 4D Convolutional Swin Transformer, where a high-dimensional Swin Transformer is preceded by a series of small-kernel convolutions that impart local context to all pixels and introduce convolutional inductive bias. We additionally boost aggregation performance by applying transformers within a pyramidal structure, where aggregation at a coarser level guides aggregation at a finer level. Noise in the transformer output is then filtered in the subsequent decoder with the help of the query's appearance embedding. With this model, a new state-of-the-art is set for all the standard benchmarks in few-shot segmentation. It is shown that VAT attains state-of-the-art performance for semantic correspondence as well, where cost aggregation also plays a central role.

Alexander Veicht, Sunghwan Hong, Dániel Baráth et al.

Feed-forward 3D Gaussian Splatting methods reconstruct a scene from posed or pose-free images in a single forward pass, yet current approaches predict one Gaussian per input pixel, tying the representation budget to camera resolution rather than scene complexity. A flat wall and a richly textured object thus produce equally many Gaussians despite very different geometric needs. We propose ZipSplat, a token-based feed-forward model that decouples Gaussian placement from the pixel grid. A multi-view backbone extracts dense visual tokens, and k-means clustering compresses them into a compact set of scene tokens. Cross- and self-attention refine these tokens, and a lightweight MLP decodes each into a group of Gaussians with unconstrained 3D positions. Because clustering is applied at inference, a single trained model spans the quality-efficiency curve without retraining. ZipSplat operates without ground-truth poses or intrinsics, yet sets a new state of the art on DL3DV and RealEstate10K with ${\sim}6{\times}$ fewer Gaussians than pixel-aligned methods, surpassing the best pose-free baseline by 2.1dB and 1.2dB PSNR, respectively. It further generalizes zero-shot to Mip-NeRF360 and ScanNet++, outperforming all comparable baselines. Our project page is at ${\href{https://veichta.com/zipsplat}{https://veichta.com/zipsplat}}$.

Sunghwan Hong, Jisu Nam, Seokju Cho et al. · nvidia, utoronto

Existing pipelines of semantic correspondence commonly include extracting high-level semantic features for the invariance against intra-class variations and background clutters. This architecture, however, inevitably results in a low-resolution matching field that additionally requires an ad-hoc interpolation process as a post-processing for converting it into a high-resolution one, certainly limiting the overall performance of matching results. To overcome this, inspired by recent success of implicit neural representation, we present a novel method for semantic correspondence, called Neural Matching Field (NeMF). However, complicacy and high-dimensionality of a 4D matching field are the major hindrances, which we propose a cost embedding network to process a coarse cost volume to use as a guidance for establishing high-precision matching field through the following fully-connected network. Nevertheless, learning a high-dimensional matching field remains challenging mainly due to computational complexity, since a naive exhaustive inference would require querying from all pixels in the 4D space to infer pixel-wise correspondences. To overcome this, we propose adequate training and inference procedures, which in the training phase, we randomly sample matching candidates and in the inference phase, we iteratively performs PatchMatch-based inference and coordinate optimization at test time. With these combined, competitive results are attained on several standard benchmarks for semantic correspondence. Code and pre-trained weights are available at https://ku-cvlab.github.io/NeMF/.

Yuanwen Yue, Damien Robert, Jianyuan Wang et al.

Modern neural architectures for 3D point cloud processing contain both convolutional layers and attention blocks, but the best way to assemble them remains unclear. We analyse the role of different computational blocks in 3D point cloud networks and find an intuitive behaviour: convolution is adequate to extract low-level geometry at high-resolution in early layers, where attention is expensive without bringing any benefits; attention captures high-level semantics and context in low-resolution, deep layers more efficiently. Guided by this design principle, we propose a new, improved 3D point cloud backbone that employs convolutions in early stages and switches to attention for deeper layers. To avoid the loss of spatial layout information when discarding redundant convolution layers, we introduce a novel, training-free 3D positional encoding, PointROPE. The resulting LitePT model has $3.6\times$ fewer parameters, runs $2\times$ faster, and uses $2\times$ less memory than the state-of-the-art Point Transformer V3, but nonetheless matches or even outperforms it on a range of tasks and datasets. Code and models are available at: https://github.com/prs-eth/LitePT.

Sunghwan Hong, Seokju Cho, Seungryong Kim et al. · nvidia, utoronto

We present a novel architecture for dense correspondence. The current state-of-the-art are Transformer-based approaches that focus on either feature descriptors or cost volume aggregation. However, they generally aggregate one or the other but not both, though joint aggregation would boost each other by providing information that one has but other lacks, i.e., structural or semantic information of an image, or pixel-wise matching similarity. In this work, we propose a novel Transformer-based network that interleaves both forms of aggregations in a way that exploits their complementary information. Specifically, we design a self-attention layer that leverages the descriptor to disambiguate the noisy cost volume and that also utilizes the cost volume to aggregate features in a manner that promotes accurate matching. A subsequent cross-attention layer performs further aggregation conditioned on the descriptors of both images and aided by the aggregated outputs of earlier layers. We further boost the performance with hierarchical processing, in which coarser level aggregations guide those at finer levels. We evaluate the effectiveness of the proposed method on dense matching tasks and achieve state-of-the-art performance on all the major benchmarks. Extensive ablation studies are also provided to validate our design choices.

Mungyeom Kim, Minkyeong Jeon, Honggyu An et al.

Dynamic scene reconstruction from monocular video remains a fundamental challenge in computer vision. Existing feed-forward methods predict 3D Gaussians pixel-wise for each frame, suffering from duplicated Gaussians and view-dependent biases that hinder effective learning of scene motion. We present C4G, a feed-forward 4D reconstruction framework built upon a compact set of timestamp-conditioned learnable Gaussian query tokens. Each token aggregates corresponding features across the full temporal context and decodes a 3D Gaussian whose position is modulated by the target timestamp, enabling globally coherent motion modeling without per-scene optimization. To capture fine-grained details, we further introduce a video diffusion model-based rendering enhancement module. Since our framework effectively aggregates features into Gaussians, we extend this capability to feature lifting, producing a 4D feature field that supports point tracking and dynamic scene understanding. C4G achieves strong novel-view synthesis performance using significantly fewer Gaussians and without requiring camera poses, while exhibiting stronger motion modeling and robustness to large temporal gaps.

Heeseong Shin, Chaehyun Kim, Sunghwan Hong et al.

Large-scale vision-language models like CLIP have demonstrated impressive open-vocabulary capabilities for image-level tasks, excelling in recognizing what objects are present. However, they struggle with pixel-level recognition tasks like semantic segmentation, which additionally require understanding where the objects are located. In this work, we propose a novel method, PixelCLIP, to adapt the CLIP image encoder for pixel-level understanding by guiding the model on where, which is achieved using unlabeled images and masks generated from vision foundation models such as SAM and DINO. To address the challenges of leveraging masks without semantic labels, we devise an online clustering algorithm using learnable class names to acquire general semantic concepts. PixelCLIP shows significant performance improvements over CLIP and competitive results compared to caption-supervised methods in open-vocabulary semantic segmentation. Project page is available at https://cvlab-kaist.github.io/PixelCLIP

Jini Yang, Eunbeen Hong, Soowon Son et al.

Event cameras capture per-pixel brightness changes with microsecond resolution, offering continuous motion information lost between RGB frames. However, existing event-based motion estimators depend on large-scale synthetic data that often suffers from a significant sim-to-real gap. We propose TETO (Tracking Events with Teacher Observation), a teacher-student framework that learns event motion estimation from only $\sim$25 minutes of unannotated real-world recordings through knowledge distillation from a pretrained RGB tracker. Our motion-aware data curation and query sampling strategy maximizes learning from limited data by disentangling object motion from dominant ego-motion. The resulting estimator jointly predicts point trajectories and dense optical flow, which we leverage as explicit motion priors to condition a pretrained video diffusion transformer for frame interpolation. We achieve state-of-the-art point tracking on EVIMO2 and optical flow on DSEC using orders of magnitude less training data, and demonstrate that accurate motion estimation translates directly to superior frame interpolation quality on BS-ERGB and HQ-EVFI.

Honggyu An, Jaewoo Jung, Mungyeom Kim et al.

Reconstructing and understanding 3D scenes from unposed sparse views in a feed-forward manner remains as a challenging task in 3D computer vision. Recent approaches use per-pixel 3D Gaussian Splatting for reconstruction, followed by a 2D-to-3D feature lifting stage for scene understanding. However, they generate excessive redundant Gaussians, causing high memory overhead and sub-optimal multi-view feature aggregation, leading to degraded novel view synthesis and scene understanding performance. We propose C3G, a novel feed-forward framework that estimates compact 3D Gaussians only at essential spatial locations, minimizing redundancy while enabling effective feature lifting. We introduce learnable tokens that aggregate multi-view features through self-attention to guide Gaussian generation, ensuring each Gaussian integrates relevant visual features across views. We then exploit the learned attention patterns for Gaussian decoding to efficiently lift features. Extensive experiments on pose-free novel view synthesis, 3D open-vocabulary segmentation, and view-invariant feature aggregation demonstrate our approach's effectiveness. Results show that a compact yet geometrically meaningful representation is sufficient for high-quality scene reconstruction and understanding, achieving superior memory efficiency and feature fidelity compared to existing methods.

Jisang Han, Sunghwan Hong, Jaewoo Jung et al.

Reliable 3D reconstruction from in-the-wild image collections is often hindered by "noisy" images-irrelevant inputs with little or no view overlap with others. While traditional Structure-from-Motion pipelines handle such cases through geometric verification and outlier rejection, feed-forward 3D reconstruction models lack these explicit mechanisms, leading to degraded performance under in-the-wild conditions. In this paper, we discover that the existing feed-forward reconstruction model, e.g., VGGT, despite lacking explicit outlier-rejection mechanisms or noise-aware training, can inherently distinguish distractor images. Through an in-depth analysis under varying proportions of synthetic distractors, we identify a specific layer that naturally exhibits outlier-suppressing behavior. Further probing reveals that this layer encodes discriminative internal representations that enable an effective noise-filtering capability, which we simply leverage to perform outlier-view rejection in feed-forward 3D reconstruction without any additional fine-tuning or supervision. Extensive experiments on both controlled and in-the-wild datasets demonstrate that this implicit filtering mechanism is consistent and generalizes well across diverse scenarios.

Sunghwan Hong, Jaewoo Jung, Heeseong Shin et al.

We consider the problem of novel view synthesis from unposed images in a single feed-forward. Our framework capitalizes on fast speed, scalability, and high-quality 3D reconstruction and view synthesis capabilities of 3DGS, where we further extend it to offer a practical solution that relaxes common assumptions such as dense image views, accurate camera poses, and substantial image overlaps. We achieve this through identifying and addressing unique challenges arising from the use of pixel-aligned 3DGS: misaligned 3D Gaussians across different views induce noisy or sparse gradients that destabilize training and hinder convergence, especially when above assumptions are not met. To mitigate this, we employ pre-trained monocular depth estimation and visual correspondence models to achieve coarse alignments of 3D Gaussians. We then introduce lightweight, learnable modules to refine depth and pose estimates from the coarse alignments, improving the quality of 3D reconstruction and novel view synthesis. Furthermore, the refined estimates are leveraged to estimate geometry confidence scores, which assess the reliability of 3D Gaussian centers and condition the prediction of Gaussian parameters accordingly. Extensive evaluations on large-scale real-world datasets demonstrate that PF3plat sets a new state-of-the-art across all benchmarks, supported by comprehensive ablation studies validating our design choices. project page: https://cvlab-kaist.github.io/PF3plat/

Sunghwan Hong, Jaewoo Jung, Heeseong Shin et al.

This work delves into the task of pose-free novel view synthesis from stereo pairs, a challenging and pioneering task in 3D vision. Our innovative framework, unlike any before, seamlessly integrates 2D correspondence matching, camera pose estimation, and NeRF rendering, fostering a synergistic enhancement of these tasks. We achieve this through designing an architecture that utilizes a shared representation, which serves as a foundation for enhanced 3D geometry understanding. Capitalizing on the inherent interplay between the tasks, our unified framework is trained end-to-end with the proposed training strategy to improve overall model accuracy. Through extensive evaluations across diverse indoor and outdoor scenes from two real-world datasets, we demonstrate that our approach achieves substantial improvement over previous methodologies, especially in scenarios characterized by extreme viewpoint changes and the absence of accurate camera poses.

Honggyu An, Jinhyeon Kim, Seonghoon Park et al.

In this work, we explore new perspectives on cross-view completion learning by drawing an analogy to self-supervised correspondence learning. Through our analysis, we demonstrate that the cross-attention map within cross-view completion models captures correspondence more effectively than other correlations derived from encoder or decoder features. We verify the effectiveness of the cross-attention map by evaluating on both zero-shot matching and learning-based geometric matching and multi-frame depth estimation. Project page is available at https://cvlab-kaist.github.io/ZeroCo/.

Sunghwan Hong, Seokju Cho, Seungryong Kim et al.

This paper introduces a Transformer-based integrative feature and cost aggregation network designed for dense matching tasks. In the context of dense matching, many works benefit from one of two forms of aggregation: feature aggregation, which pertains to the alignment of similar features, or cost aggregation, a procedure aimed at instilling coherence in the flow estimates across neighboring pixels. In this work, we first show that feature aggregation and cost aggregation exhibit distinct characteristics and reveal the potential for substantial benefits stemming from the judicious use of both aggregation processes. We then introduce a simple yet effective architecture that harnesses self- and cross-attention mechanisms to show that our approach unifies feature aggregation and cost aggregation and effectively harnesses the strengths of both techniques. Within the proposed attention layers, the features and cost volume both complement each other, and the attention layers are interleaved through a coarse-to-fine design to further promote accurate correspondence estimation. Finally at inference, our network produces multi-scale predictions, computes their confidence scores, and selects the most confident flow for final prediction. Our framework is evaluated on standard benchmarks for semantic matching, and also applied to geometric matching, where we show that our approach achieves significant improvements compared to existing methods.

Jisang Han, Honggyu An, Jaewoo Jung et al.

In this work, we address the task of 3D reconstruction in dynamic scenes, where object motions frequently degrade the quality of previous 3D pointmap regression methods, such as DUSt3R, that are originally designed for static 3D scene reconstruction. Although these methods provide an elegant and powerful solution in static settings, they struggle in the presence of dynamic motions that disrupt alignment based solely on camera poses. To overcome this, we propose $D^2USt3R$ that directly regresses Static-Dynamic Aligned Pointmaps (SDAP) that simultaneiously capture both static and dynamic 3D scene geometry. By explicitly incorporating both spatial and temporal aspects, our approach successfully encapsulates 3D dense correspondence to the proposed pointmaps, enhancing downstream tasks. Extensive experimental evaluations demonstrate that our proposed approach consistently achieves superior 3D reconstruction performance across various datasets featuring complex motions.

Marcel Gröpl, Jaewoo Jung, Seungryong Kim et al.

Despite rapid progress, pretrained vision-language models still struggle when answers depend on tiny visual details or on combining clues spread across multiple regions, as in documents and compositional queries. We address this by framing grounding as test-time evidence retrieval: given a query, the model should actively identify where to look next to resolve ambiguity. To this end, we propose a training-free, model-intrinsic grounding method that uses uncertainty as supervision. Specifically, we compute the entropy of the model's next-token distribution and backpropagate it to the visual token embeddings to obtain an entropy-gradient relevance map, without auxiliary detectors or attention-map heuristics. We then extract and rank multiple coherent regions to support multi-evidence queries, and introduce an iterative zoom-and-reground procedure with a spatial-entropy stopping rule to avoid over-refinement. Experiments on seven benchmarks across four VLM architectures demonstrate consistent improvements over existing methods, with the largest gains on detail-critical and high-resolution settings, while also producing more interpretable evidence localizations.

Nahyuk Lee, Zhiang Chen, Marc Pollefeys et al.

Flow-matching methods for 3D shape assembly learn point-wise velocity fields that transport parts toward assembled configurations, yet they receive no explicit guidance about which cross-part interactions should drive the motion. We introduce TORA, a topology-first representation alignment framework that distills relational structure from a frozen pretrained 3D encoder into the flow-matching backbone during training. We first realize this via simple instantiation, token-wise cosine matching, which injects the learned geometric descriptors from the teacher representation. We then extend to employ a Centered Kernel Alignment (CKA) loss to match the similarity structure between student and teacher representations for enhanced topological alignment. Through systematic probing of diverse 3D encoders, we show that geometry- and contact-centric teacher properties, not semantic classification ability, govern alignment effectiveness, and that alignment is most beneficial at later transformer layers where spatial structure naturally emerges. TORA introduces zero inference overhead while yielding two consistent benefits: faster convergence (up to 6.9$\times$) and improved accuracy in-distribution, along with greater robustness under domain shift. Experiments on five benchmarks spanning geometric, semantic, and inter-object assembly demonstrate state-of-the-art performance, with particularly pronounced gains in zero-shot transfer to unseen real-world and synthetic datasets. Project page: https://nahyuklee.github.io/tora.

Heeji Yoon, Jaewoo Jung, Junwan Kim et al.

Multimodal large language models (MLLMs) trained with visual instruction tuning have achieved strong performance across diverse tasks, yet they remain limited in vision-centric tasks such as object counting or spatial reasoning. We attribute this gap to the prevailing text-only supervision paradigm, which provides only indirect guidance for the visual pathway and often leads MLLMs to discard fine-grained visual details during training. In this paper, we present VIsual Representation ALignment (VIRAL), a simple yet effective regularization strategy that aligns the internal visual representations of MLLMs with those of pre-trained vision foundation models (VFMs). By explicitly enforcing this alignment, VIRAL enables the model not only to retain critical visual details from the input vision encoder but also to complement additional visual knowledge from VFMs, thereby enhancing its ability to reason over complex visual inputs. Our experiments demonstrate consistent improvements across all tasks on widely adopted multimodal benchmarks. Furthermore, we conduct comprehensive ablation studies to validate the key design choices underlying our framework. We believe this simple finding opens up an important direction for the effective integration of visual information in training MLLMs.

A. Said Gurbuz, Sunghwan Hong, Ahmed Nassar et al.

Modern computer-use agents (CUA) must perceive a screen as a structured state, what elements are visible, where they are, and what text they contain, before they can reliably ground instructions and act. Yet, most available grounding datasets provide sparse supervision, with insufficient and low-diversity labels that annotate only a small subset of task-relevant elements per screen, which limits both coverage and generalization; moreover, practical deployment requires efficiency to enable low-latency, on-device use. We introduce ScreenParse, a large-scale dataset for complete screen parsing, with dense annotations of all visible UI elements (boxes, 55-class types, and text) across 771K web screenshots (21M elements). ScreenParse is generated by Webshot, an automated, scalable pipeline that renders diverse urls, extracts annotations and applies VLM-based relabeling and quality filtering. Using ScreenParse, we train ScreenVLM, a compact, 316M-parameter vision language model (VLM) that decodes a compact ScreenTag markup representation with a structure-aware loss that upweights structure-critical tokens. ScreenVLM substantially outperforms much larger foundation VLMs on dense parsing (e.g., 0.592 vs. 0.294 PageIoU on ScreenParse) and shows strong transfer to public benchmarks. Moreover, finetuning foundation VLMs on ScreenParse consistently improves their grounding performance, suggesting that dense screen supervision provides transferable structural priors for UI understanding. Project page: https://saidgurbuz.github.io/screenparse/.

Chaehyun Kim, Heeseong Shin, Eunbeen Hong et al.

Text-to-image diffusion models excel at translating language prompts into photorealistic images by implicitly grounding textual concepts through their cross-modal attention mechanisms. Recent multi-modal diffusion transformers extend this by introducing joint self-attention over concatenated image and text tokens, enabling richer and more scalable cross-modal alignment. However, a detailed understanding of how and where these attention maps contribute to image generation remains limited. In this paper, we introduce Seg4Diff (Segmentation for Diffusion), a systematic framework for analyzing the attention structures of MM-DiT, with a focus on how specific layers propagate semantic information from text to image. Through comprehensive analysis, we identify a semantic grounding expert layer, a specific MM-DiT block that consistently aligns text tokens with spatially coherent image regions, naturally producing high-quality semantic segmentation masks. We further demonstrate that applying a lightweight fine-tuning scheme with mask-annotated image data enhances the semantic grouping capabilities of these layers and thereby improves both segmentation performance and generated image fidelity. Our findings demonstrate that semantic grouping is an emergent property of diffusion transformers and can be selectively amplified to advance both segmentation and generation performance, paving the way for unified models that bridge visual perception and generation.

JoungBin Lee, Jaewoo Jung, Jisang Han et al.

We present 3DScenePrompt, a framework that generates the next video chunk from arbitrary-length input while enabling precise camera control and preserving scene consistency. Unlike methods conditioned on a single image or a short clip, we employ dual spatio-temporal conditioning that reformulates context-view referencing across the input video. Our approach conditions on both temporally adjacent frames for motion continuity and spatially adjacent content for scene consistency. However, when generating beyond temporal boundaries, directly using spatially adjacent frames would incorrectly preserve dynamic elements from the past. We address this by introducing a 3D scene memory that represents exclusively the static geometry extracted from the entire input video. To construct this memory, we leverage dynamic SLAM with our newly introduced dynamic masking strategy that explicitly separates static scene geometry from moving elements. The static scene representation can then be projected to any target viewpoint, providing geometrically consistent warped views that serve as strong 3D spatial prompts while allowing dynamic regions to evolve naturally from temporal context. This enables our model to maintain long-range spatial coherence and precise camera control without sacrificing computational efficiency or motion realism. Extensive experiments demonstrate that our framework significantly outperforms existing methods in scene consistency, camera controllability, and generation quality. Project page : https://cvlab-kaist.github.io/3DScenePrompt/

Susung Hong, Junyoung Seo, Heeseong Shin et al.

In the paradigm of AI-generated content (AIGC), there has been increasing attention to transferring knowledge from pre-trained text-to-image (T2I) models to text-to-video (T2V) generation. Despite their effectiveness, these frameworks face challenges in maintaining consistent narratives and handling shifts in scene composition or object placement from a single abstract user prompt. Exploring the ability of large language models (LLMs) to generate time-dependent, frame-by-frame prompts, this paper introduces a new framework, dubbed DirecT2V. DirecT2V leverages instruction-tuned LLMs as directors, enabling the inclusion of time-varying content and facilitating consistent video generation. To maintain temporal consistency and prevent mapping the value to a different object, we equip a diffusion model with a novel value mapping method and dual-softmax filtering, which do not require any additional training. The experimental results validate the effectiveness of our framework in producing visually coherent and storyful videos from abstract user prompts, successfully addressing the challenges of zero-shot video generation.

Seokju Cho, Sunghwan Hong, Seungryong Kim

Cost aggregation is a highly important process in image matching tasks, which aims to disambiguate the noisy matching scores. Existing methods generally tackle this by hand-crafted or CNN-based methods, which either lack robustness to severe deformations or inherit the limitation of CNNs that fail to discriminate incorrect matches due to limited receptive fields and inadaptability. In this paper, we introduce Cost Aggregation with Transformers (CATs) to tackle this by exploring global consensus among initial correlation map with the help of some architectural designs that allow us to fully enjoy global receptive fields of self-attention mechanism. Also, to alleviate some of the limitations that CATs may face, i.e., high computational costs induced by the use of a standard transformer that its complexity grows with the size of spatial and feature dimensions, which restrict its applicability only at limited resolution and result in rather limited performance, we propose CATs++, an extension of CATs. Our proposed methods outperform the previous state-of-the-art methods by large margins, setting a new state-of-the-art for all the benchmarks, including PF-WILLOW, PF-PASCAL, and SPair-71k. We further provide extensive ablation studies and analyses.

Sunghwan Hong, Seokju Cho, Jisu Nam et al.

We introduce a novel cost aggregation network, dubbed Volumetric Aggregation with Transformers (VAT), to tackle the few-shot segmentation task by using both convolutions and transformers to efficiently handle high dimensional correlation maps between query and support. In specific, we propose our encoder consisting of volume embedding module to not only transform the correlation maps into more tractable size but also inject some convolutional inductive bias and volumetric transformer module for the cost aggregation. Our encoder has a pyramidal structure to let the coarser level aggregation to guide the finer level and enforce to learn complementary matching scores. We then feed the output into our affinity-aware decoder along with the projected feature maps for guiding the segmentation process. Combining these components, we conduct experiments to demonstrate the effectiveness of the proposed method, and our method sets a new state-of-the-art for all the standard benchmarks in few-shot segmentation task. Furthermore, we find that the proposed method attains state-of-the-art performance even for the standard benchmarks in semantic correspondence task although not specifically designed for this task. We also provide an extensive ablation study to validate our architectural choices. The trained weights and codes are available at: https://seokju-cho.github.io/VAT/.

Hojoon Lee, Dongyoon Hwang, Sunghwan Hong et al.

Successful sequential recommendation systems rely on accurately capturing the user's short-term and long-term interest. Although Transformer-based models achieved state-of-the-art performance in the sequential recommendation task, they generally require quadratic memory and time complexity to the sequence length, making it difficult to extract the long-term interest of users. On the other hand, Multi-Layer Perceptrons (MLP)-based models, renowned for their linear memory and time complexity, have recently shown competitive results compared to Transformer in various tasks. Given the availability of a massive amount of the user's behavior history, the linear memory and time complexity of MLP-based models make them a promising alternative to explore in the sequential recommendation task. To this end, we adopted MLP-based models in sequential recommendation but consistently observed that MLP-based methods obtain lower performance than those of Transformer despite their computational benefits. From experiments, we observed that introducing explicit high-order interactions to MLP layers mitigates such performance gap. In response, we propose the Multi-Order Interaction (MOI) layer, which is capable of expressing an arbitrary order of interactions within the inputs while maintaining the memory and time complexity of the MLP layer. By replacing the MLP layer with the MOI layer, our model was able to achieve comparable performance with Transformer-based models while retaining the MLP-based models' computational benefits.

Sunghwan Hong, Seungryong Kim

Conventional techniques to establish dense correspondences across visually or semantically similar images focused on designing a task-specific matching prior, which is difficult to model. To overcome this, recent learning-based methods have attempted to learn a good matching prior within a model itself on large training data. The performance improvement was apparent, but the need for sufficient training data and intensive learning hinders their applicability. Moreover, using the fixed model at test time does not account for the fact that a pair of images may require their own prior, thus providing limited performance and poor generalization to unseen images. In this paper, we show that an image pair-specific prior can be captured by solely optimizing the untrained matching networks on an input pair of images. Tailored for such test-time optimization for dense correspondence, we present a residual matching network and a confidence-aware contrastive loss to guarantee a meaningful convergence. Experiments demonstrate that our framework, dubbed Deep Matching Prior (DMP), is competitive, or even outperforms, against the latest learning-based methods on several benchmarks for geometric matching and semantic matching, even though it requires neither large training data nor intensive learning. With the networks pre-trained, DMP attains state-of-the-art performance on all benchmarks.

Seokju Cho, Sunghwan Hong, Sangryul Jeon et al.

We propose a novel cost aggregation network, called Cost Aggregation Transformers (CATs), to find dense correspondences between semantically similar images with additional challenges posed by large intra-class appearance and geometric variations. Cost aggregation is a highly important process in matching tasks, which the matching accuracy depends on the quality of its output. Compared to hand-crafted or CNN-based methods addressing the cost aggregation, in that either lacks robustness to severe deformations or inherit the limitation of CNNs that fail to discriminate incorrect matches due to limited receptive fields, CATs explore global consensus among initial correlation map with the help of some architectural designs that allow us to fully leverage self-attention mechanism. Specifically, we include appearance affinity modeling to aid the cost aggregation process in order to disambiguate the noisy initial correlation maps and propose multi-level aggregation to efficiently capture different semantics from hierarchical feature representations. We then combine with swapping self-attention technique and residual connections not only to enforce consistent matching but also to ease the learning process, which we find that these result in an apparent performance boost. We conduct experiments to demonstrate the effectiveness of the proposed model over the latest methods and provide extensive ablation studies. Project page is available at : https://sunghwanhong.github.io/CATs/.