TF-SSD: A Strong Pipeline via Synergic Mask Filter for Training-free Co-salient Object Detection

Zhijin He; Shuo Jin; Siyue Yu; Shuwei Wu; Bingfeng Zhang; Li Yu; Jimin Xiao

cs.CVApr 1, 2026

TF-SSD: A Strong Pipeline via Synergic Mask Filter for Training-free Co-salient Object Detection

Authors: Zhijin He, Shuo Jin, Siyue Yu, Shuwei Wu, Bingfeng Zhang, Li Yu, Jimin Xiao

Organizations: XJTLU · University of Liverpool · China University of Petroleum (East China) · Nanjing University of Information Science and Technology

Abstract

Co-salient Object Detection (CoSOD) aims to segment salient objects that consistently appear across a group of related images. Despite the notable progress achieved by recent training-based approaches, they still remain constrained by the closed-set datasets and exhibit limited generalization. However, few studies explore the potential of Vision Foundation Models (VFMs) to address CoSOD, which demonstrate a strong generalized ability and robust saliency understanding. In this paper, we investigate and leverage VFMs for CoSOD, and further propose a novel training-free method, TF-SSD, through the synergy between SAM and DINO. Specifically, we first utilize SAM to generate comprehensive raw proposals, which serve as a candidate mask pool. Then, we introduce a quality mask generator to filter out redundant masks, thereby acquiring a refined mask set. Since this generator is built upon SAM, it inherently lacks semantic understanding of saliency. To this end, we adopt an intra-image saliency filter that employs DINO's attention maps to identify visually salient masks within individual images. Moreover, to extend saliency understanding across group images, we propose an inter-image prototype selector, which computes similarity scores among cross-image prototypes to select masks with the highest score. These selected masks serve as final predictions for CoSOD. Extensive experiments show that our TF-SSD outperforms existing methods (e.g., 13.7% gains over the recent training-free method). Codes are available at https://github.com/hzz-yy/TF-SSD.

Figures & tables

Figure 1: Segmentation results derived from SAM. The oracle masks selected under the ground truth guidance exhibit co-salient representations, which are highlighted in the green rectangle.

Figure 2: Comparison of our TF-SSD and other CoSOD methods. $F_{m}$ and $S_{m}$ denote the F-measure and S-measure, respectively. Our method surpasses training-based methods and achieves 13.7% improvement over the recent training-free SOTA method.

Figure 3: Overview pipeline of our TF-SSD. It contains four components: a Quality Mask Generator (QMG) that filters out exhaustive candidate masks from SAM, a DINO encoder for salient attention and semantic prototype extraction, an Intra-image Saliency Filter (ISF) for visually salient purification within a single image, and an Inter-image Prototype Selector (IPS) that builds the saliency relationship across group images to discover the co-salient masks as the final predictions for CoSOD. For clarity, our notations are defined as: $m_{n,t}$ denotes an individual mask, $\textbf{M}_{n}$ denotes the mask set of image $n$ (e.g., $\textbf{M}_{n}^{raw}$ , $\textbf{M}_{n}^{refine}$ ).

Panel of Figure 4 — Algorithm 1 Quality Mask Generator (QMG)

Figure 4: Illustration of Intra-image Saliency Filter. Row 1 : Original image, self-attention map of DINO, and ground truth. Row 2 : Three mask proposals generated by QMG. Row 3 : Attention response obtained by element-wise multiplication of each mask with the attention map. Mask 1 achieves a high saliency score due to strong overlap between salient regions and the mask proposal, while Masks 2 and 3 receive low scores due to non-overlap.

Figure 5: Illustration of Inter-image Prototype Selector (IPS) with an example of N=3 images and T=3 masks per image. A pairwise similarity matrix is constructed using mask prototypes, where different colors denote prototypes from different images. For each candidate (e.g., $p_{1,1}$ of image $I_{1}$ ), we select its maximum score (deep red cells) against the candidates from each of the other images ( $I_{2}$ and $I_{3}$ ). These maximum scores are summed as the total co-saliency score $s^{co}_{max}$ . The mask with the highest $s^{co}_{max}$ is selected as the final prediction.

	Component					CoCA [ 44 ]				CoSal2015 [ 41 ]				CoSOD3k [ 8 ]
ID	QMG-1	QMG-2	QMG-3	ISF	IPS	$MAE\downarrow$	$F_{\beta}^{max}\uparrow$	$E_{\xi}^{max}\uparrow$	$S_{\alpha}\uparrow$	$MAE\downarrow$	$F_{\beta}^{max}\uparrow$	$E_{\xi}^{max}\uparrow$	$S_{\alpha}\uparrow$	$MAE\downarrow$	$F_{\beta}^{max}\uparrow$	$E_{\xi}^{max}\uparrow$	$S_{\alpha}\uparrow$
1	✓				✓	0.171	0.447	0.662	0.578	0.133	0.680	0.779	0.704	0.139	0.610	0.752	0.667
2	✓	✓			✓	0.121	0.497	0.747	0.650	0.095	0.815	0.875	0.799	0.099	0.744	0.839	0.754
3	✓	✓	✓		✓	0.119	0.522	0.765	0.634	0.092	0.850	0.900	0.818	0.099	0.785	0.862	0.786
5	✓	✓	✓	✓	✓	0.077	0.686	0.815	0.763	0.089	0.899	0.926	0.890	0.090	0.860	0.908	0.861

Table 2: Ablation study of the proposed components in TF-SSD on three benchmark datasets. QMG-1, QMG-2, and QMG-3 denote the three stages of our QMG, respectively. The final row represents our full framework. Best results are marked bold .

Figure 6: Qualitative comparison. We compare the result visualization between our method and previous SOTA methods. Among them, ZS-CoSOD [ 34 ] is the training-free method. GCoNet+ [ 45 ] , VCP [ 31 ] and SCoSPARC [ 3 ] are training-based methods. It’s observed that TF-SSD obtains precise co-salient masks compared to other SOTA methods, even in complicated cases ( e.g . , “Chopsticks” and “Axe”).

Strategy	CoCA		CoSal2015		CoSOD3k
Strategy	$F_{\beta}^{\text{max}}\uparrow$	$E_{\xi}^{\text{max}}\uparrow$	$F_{\beta}^{\text{max}}\uparrow$	$E_{\xi}^{\text{max}}\uparrow$	$F_{\beta}^{\text{max}}\uparrow$	$E_{\xi}^{\text{max}}\uparrow$
Avg	0.568	0.778	0.775	0.826	0.737	0.792
Max	0.686	0.815	0.899	0.926	0.860	0.908

Table 3: Ablation study of the

N-1

similarity score selection (Eq. 9 ) in IPS. We compare using the maximum similarity ( Max ) versus the average similarity ( Avg ) to compute the similarity score among cross-image prototypes.

T	MAE $\downarrow$	$F_{\beta}^{\text{ max}}\uparrow$	$E_{\xi}^{\text{max}}\uparrow$	$S_{\alpha}\uparrow$
10	0.078	0.616	0.786	0.718
8	0.078	0.645	0.789	0.726
7	0.077	0.666	0.809	0.744
6	0.077	0.686	0.815	0.763
5	0.077	0.667	0.804	0.748

Table 4: Ablation study on the number of selected masks (

T

) in the ISF module . Performance is evaluated on the CoCA dataset.

$\tau_{area}$	$F_{\beta}^{\text{max}}$	Oracle $F_{\beta}^{\text{max}}$
0.002	0.545	0.696
0.005	0.601	0.722
0.01	0.686	0.768
0.02	0.653	0.737

Table 5: Ablation study on different area ratio thresholds

\tau_{area}

on the CoCA dataset.

Backbone	$F_{\beta}^{\text{max}}$	$E_{\xi}^{\text{max}}$	$S_{\alpha}$
DINO ViT-B/16	0.556	0.742	0.697
DINOv2 ViT-B/14	0.607	0.783	0.708
DINO ViT-B/8	0.686	0.815	0.763

Table 6: Performance comparison of different DINO backbones on the CoCA dataset.

$\alpha$	$\beta$	$F_{\beta}^{\text{max}}$	$E_{\xi}^{\text{max}}$	$S_{\alpha}$
0.0	1.0	0.473	0.702	0.611
0.3	0.7	0.591	0.737	0.651
0.5	0.5	0.658	0.795	0.744
0.7	0.3	0.686	0.815	0.763
1.0	0.0	0.641	0.782	0.735

Table 7: Ablation on quality score weights on the CoCA dataset.

$r_{min}$	$r_{max}$	$F_{\beta}^{\text{max}}$	$E_{\xi}^{\text{max}}$	$S_{\alpha}$
0.05	0.85	0.661	0.798	0.747
0.1	0.6	0.678	0.808	0.756
0.15	0.7	0.686	0.815	0.763
0.2	0.5	0.619	0.741	0.669

Table 8: Ablation on ideal size range on the CoCA dataset.

$\tau_{con}$	$F_{\beta}^{\text{max}}$	$E_{\xi}^{\text{max}}$	$S_{\alpha}$
0.5	0.652	0.791	0.741
0.7	0.673	0.805	0.754
0.85	0.686	0.815	0.763
0.95	0.679	0.796	0.756

Table 9: Ablation on overlap threshold on the CoCA dataset.

Figure 7: Visualization of SAM mask proposals for ”Accordion” and ”bicycle” image groups. For each group, we show the original images (row 1), ground truth(GT) masks (row 2), and the top-5 mask proposals ranked by quality score $S_{n,t}^{ba}$ (rows 3-7). Green borders indicate the final predictions selected by our TF-SSD pipeline.

Figure 8: Visualization of DINO attention maps for three categories: ”airplane”, ”butterfly”, and ”penguin”. Row 1: Resized input images. Row 2: Resized GT masks. Row 3: DINO attention maps $\mathcal{A}_{n}$ that naturally highlight salient objects.

Figure 9: Limitation on small object detection. Visualizations of failed cases from the ”moon” category.

Salient Object Detection

Salient Object Detection finds visually prominent objects in images or video and predicts their pixel-level foreground masks.

See all 37 papers

Rank-Consistent Set Reasoning for Co-Salient Object Detection
Sep 12, 2026Yuan Xiang, Matteo Rossi, Yingzhou ChenSalient Object Detection
University of California, Los Angeles (UCLA) USA · Polytechnic University of Turin Italy
Global-Local Feature Decoding with Adapter-Guided SAMv2 for Salient Object Detection
May 4, 2026Morteza Moradi, Mohammad Moradi, Simone Palazzo +2Salient Object Detection Vision Foundation Model Adaptation
University of Catania, Catania, Italy · Quintic AI, San Francisco, CA, USA
GFR-SAM: Training-Free Referring Camouflaged Object Segmentation via Cross-Image Prompting
Jul 13, 2026Yilong Yang, Jianxin Tian, Shengchuan Zhang +1Image Segmentation Promptable Image Segmentation
Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University

Sep 12, 2026J/K move · Enter open · S save

Rank-Consistent Set Reasoning for Co-Salient Object Detection

Yuan Xiang, Matteo Rossi, Yingzhou Chen

University of California, Los Angeles (UCLA) USA · Polytechnic University of Turin Italy

Salient Object Detection

Open paper AskPDF arXiv

Co-salient object detection (Co-SOD) requires a model to find foreground regions that are salient in individual images and supported by the image group. We present Rank-Consistent Set Reasoning (RCSR), a supervised dense-prediction framework that models a group as an unordered set rather than as a sequence of images or a semantic label. The core idea is to rank how strongly each spatial region agrees with a small collection of learned group slots at every image scale, and to aggregate these ranks with a robust trimmed statistic. This suppresses accidental pairwise matches and prevents one atypical group member from dominating the shared representation. A set encoder builds group slots directly from multi-scale visual features, while a rank-consistency gate measures whether the ordering of candidate regions is stable across group members. The gated slots are decoded jointly with per-image features to produce co-saliency maps. The model contains no natural-language branch, no open-vocabulary detector, and no external segmentation model. We further introduce a group permutation objective and hard-distractor augmentation so that the model learns the properties of a set-level target rather than memorizing image order or isolated visual saliency. We formulate an evaluation protocol for CoCA, CoSal2015, and CoSOD3k, together with tests of group-size robustness, distractor rejection, order invariance, and cross-dataset transfer.

Abstract

Salient Object Detection

Explore similar work

Rank-Consistent Set Reasoning for Co-Salient Object Detection

Global-Local Feature Decoding with Adapter-Guided SAMv2 for Salient Object Detection

GFR-SAM: Training-Free Referring Camouflaged Object Segmentation via Cross-Image Prompting