CatalogStitch: Dimension-Aware and Occlusion-Preserving Object Compositing for Catalog Image Generation

Sanyam Jain; Pragya Kandari; Manit Singhal; He Zhang; Soo Ye Kim

cs.CVApr 10, 2026

CatalogStitch: Dimension-Aware and Occlusion-Preserving Object Compositing for Catalog Image Generation

Authors: Sanyam Jain, Pragya Kandari, Manit Singhal, He Zhang, Soo Ye Kim

Organizations: Adobe

Abstract

Generative object compositing methods have shown remarkable ability to seamlessly insert objects into scenes. However, when applied to real-world catalog image generation, these methods require tedious manual intervention: users must carefully adjust masks when product dimensions differ, and painstakingly restore occluded elements post-generation. We present CatalogStitch, a set of model-agnostic techniques that automate these corrections, enabling user-friendly content creation. Our dimension-aware mask computation algorithm automatically adapts the target region to accommodate products with different dimensions; users simply provide a product image and background, without manual mask adjustments. Our occlusion-aware hybrid restoration method guarantees pixel-perfect preservation of occluding elements, eliminating post-editing workflows. We additionally introduce CatalogStitch-Eval, a 58-example benchmark covering aspect-ratio mismatch and occlusion-heavy catalog scenarios, together with supplementary PDF and HTML viewers. We evaluate our techniques with three state-of-the-art compositing models (ObjectStitch, OmniPaint, and InsertAnything), demonstrating consistent improvements across diverse catalog scenarios. By reducing manual intervention and automating tedious corrections, our approach transforms generative compositing into a practical, human-friendly tool for production catalog workflows.

Figures & tables

Panel of Figure 1 — Figure 1: Challenge 1: Product Dimension Mismatch. When replacing a product with different proportions, freeform and bounding box masks distort the product shape. Our dimension-aware mask preserves correct proportions.

Panel of Figure 2 — Figure 2: Challenge 2: Occlusion Destruction. Foreground elements (table/decor items) occluding the target product are corrupted during compositing. Our hybrid restoration preserves them pixel-perfectly.

Panel of Figure 3 — Figure 3: CatalogStitch overview. Two lightweight, model-agnostic modules wrap a baseline compositor. The target mask is adapted to the replacement product ratio; occluders are segmented, cached, and inpainted away before compositing, then restored from the original pixels at the final step.

Panel of Figure 4 — Figure 4: Module-level flow charts. Left: dimension-aware mask computation preserves the original placement when the mismatch is small and otherwise expands the target region around the original centroid. Right: occlusion-aware restoration detects overlapping entities, caches their exact pixels, applies generative inpainting to remove occluders and expose a clean background, runs the compositor on the clean background, and finally restores the cached occluder pixels to preserve original geometry, texture, and scene-consistent lighting.

Panel of Figure 5 — Figure 5: Dimension-aware mask computation across models. Left inputs: background and product. Mask overlay visualizes the dim-aware (blue) and bounding-box (pink) regions. Right: outputs under Freeform, BBox, and Dim-Aware masks for OmniPaint, InsertAnything, and ObjectStitch. Dim-aware masks preserve object proportions more reliably than Freeform and BBox masks across all models.

Panel of Figure 6 — Figure 6: Occlusion-aware hybrid restoration across models. Left inputs: background and product. Mask overlay and occluder composite are visualizations of the occlusion regions. Right: Freeform, BBox, and Dim-Aware outcomes for InsertAnything and ObjectStitch, shown before and after occluder restoration. Our pipeline segments occluders, caches their pixels, inpaints them away to expose a clean background for compositing, and finally pastes the cached pixels back—preserving overlapping structures more faithfully than the corresponding before-restoration outputs.

Method	AR Error $\downarrow$	Occluder PSNR $\uparrow$	FID $\downarrow$	CLIP-score $\uparrow$	DINO-score $\uparrow$
OmniPaint	31.07	-	137.15	83.03	68.50
OmniPaint + Ours	4.57	-	135.79	86.11	72.99
ObjectStitch	30.97	11.60	101.55	90.27	85.76
ObjectStitch + Ours	5.05	26.84	91.52	90.62	88.09
InsertAnything	29.98	13.33	105.99	90.23	82.63
InsertAnything + Ours	3.92	27.54	77.72	92.68	88.30

Table 1: Quantitative comparison on CatalogStitch-Eval. AR Error, CLIP, and DINO scores are computed on the 35 dimension examples; Occluder PSNR on the 23 occlusion examples (– for OmniPaint). CLIP and DINO are cosine similarity

\times 100

; FID is between input-product and generated-object crop distributions.

Image Editing

Image editing changes selected visual content while preserving the parts of an image that should remain unchanged.

See all 315 papers

PICS: Pairwise Image Compositing with Spatial Interactions
Mar 6, 2026Hang Zhou, Xinxin Zuo, Sen Wang +1Image Editing Compositional Image Generation
University of Alberta · Concordia University
OSInsert: Towards High-authenticity and High-fidelity Image Composition
Feb 23, 2026Jingyuan Wang, Li NiuCompositional Image Generation
MoE Key Lab of Artificial Intelligence, Shanghai Jiao Tong University
Bridging the Catalog-to-Real Gap: Scalable Product Recognition via Multi-Stage Contrastive Learning
Jul 10, 2026Anyi Zhang, Joy Mazumder, Kiril LomakinContrastive Learning Visual Search
Walmart Global Tech Sunnyvale, CA, USA

Mar 6, 2026J/K move · Enter open · S save

PICS: Pairwise Image Compositing with Spatial Interactions

Hang Zhou, Xinxin Zuo, Sen Wang, Li Cheng

University of Alberta · Concordia University

Image Editing Compositional Image Generation

Open paper AskPDF arXiv

Despite strong single-turn performance, diffusion-based image compositing often struggles to preserve coherent spatial relations in pairwise or sequential edits, where subsequent insertions may overwrite previously generated content and disrupt physical consistency. We introduce PICS, a self-supervised composition-by-decomposition paradigm that composes objects in parallel while explicitly modeling the compositional interactions among (fully-/partially-)visible objects and background. At its core, an Interaction Transformer employs mask-guided Mixture-of-Experts to route background, exclusive, and overlap regions to dedicated experts, with an adaptive α-blending strategy that infers a compatibility-aware fusion of overlapping objects while preserving boundary fidelity. To further enhance robustness to geometric variations, we incorporate geometry-aware augmentations covering both out-of-plane and in-plane pose changes of objects. Our method delivers superior pairwise compositing quality and substantially improved stability, with extensive evaluations across virtual try-on, indoor, and street scene settings showing consistent gains over state-of-the-art baselines. Code and data are available at https://github.com/RyanHangZhou/PICS

Abstract

Image Editing

Explore similar work

PICS: Pairwise Image Compositing with Spatial Interactions

OSInsert: Towards High-authenticity and High-fidelity Image Composition

Bridging the Catalog-to-Real Gap: Scalable Product Recognition via Multi-Stage Contrastive Learning